INDEX

Explanations

numerical codes or identifiers related to data entries

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

aarrggbb

-0.43

 Chwiliwch

-0.41

 للاسماء

-0.40

########.

-0.36

isContained

-0.36

 BoxDecoration

-0.35

 embarazada

-0.32

 defaultstate

-0.32

:]:

-0.32

Hentet

-0.31

POSITIVE LOGITS

UseVisualStyle

0.56

parsedMessage

0.55

mybatisplus

0.54

RegressionTest

0.54

 ModelExpression

0.54

windowFixed

0.48

UserScript

0.47

 betweenstory

0.47

chure

0.45

')")

0.44

Activations Density 0.015%

numerical codes or identifiers related to data entries

No Comments

No Known Activations