INDEX

Explanations

punctuation marks and numbers indicating quantities or lists

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.23.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

ValueStyle

-0.86

 EconPapers

-0.80

يميديا

-0.80

setViewportView

-0.79

PerformLayout

-0.79

rungsseite

-0.77

 Chwiliwch

-0.76

__':

-0.74

 autorytatywna

-0.72

 Audiodateien

-0.72

POSITIVE LOGITS

 stages

0.47

 stage

0.44

final

0.42

 phase

0.42

 final

0.41

 steps

0.41

0.40

phase

0.40

 rapid

0.39

 fast

0.39

Activations Density 0.001%

punctuation marks and numbers indicating quantities or lists

No Comments

No Known Activations