INDEX

Explanations

indicators of official or formal communication contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

-0.88

-0.86

StoryboardSegue

-0.85

-0.75

 Paglinawan

-0.72

-0.68

ьаж

-0.65

させていただきました

-0.65

 Walkover

-0.65

EndGlobalSection

-0.64

POSITIVE LOGITS

$_"

0.82

0.76

 pleaſure

0.74

enumi

0.69

 Diſ

0.67

 leaſt

0.67

&_

0.66

 ſever

0.65

neſs

0.65

Activations Density 0.236%

indicators of official or formal communication contexts

No Comments

No Known Activations