INDEX

Explanations

references to changes or modifications within a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

mius

-0.56

-0.54

get

-0.54

st

-0.54

mo

-0.53

 kadang

-0.53

zeti

-0.53

so

-0.53

そもそも

-0.53

-0.51

POSITIVE LOGITS

 changes

1.55

Changes

1.46

 Changes

1.46

changes

1.39

CHANGES

1.32

 CHANGES

1.29

 itſelf

1.13

 improvements

1.13

 amendments

1.11

 changements

1.09

Activations Density 0.280%

references to changes or modifications within a context

No Comments

No Known Activations