INDEX

Explanations

statements or answers that indicate agreement or affirmation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Speaker identification markers in transcript format, particularly "A" for answers in legal testimony and speaker names followed by colons in formal documented dialogue.

eleuther_acts_top20 · claude-3-7-sonnet-20250219 Triggered by @lu-christina

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

parsedMessage

-0.85

 desmotivaciones

-0.67

MLLoader

-0.65

 miniaturka

-0.64

imetric

-0.63

 queſta

-0.63

 MessageBoxIcon

-0.63

ésultats

-0.62

 efficacement

-0.62

ligently

-0.61

POSITIVE LOGITS

Yeah

0.29

0.28

AP

0.27

subpackage

0.27

tgz

0.26

\,

0.26

OK

0.25

 AspNetCore

0.25

Activations Density 0.005%

statements or answers that indicate agreement or affirmation

Speaker identification markers in transcript format, particularly "A" for answers in legal testimony and speaker names followed by colons in formal documented dialogue.

No Comments

No Known Activations