INDEX

Explanations

verbs indicating actions, processes, or changes

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

المشاركات

-0.61

Couldn

-0.57

<eos>

-0.56

-0.53

 couldn

-0.52

 would

-0.49

which

-0.48

was

-0.47

would

-0.46

POSITIVE LOGITS

 fhew

1.01

 Theſe

0.99

AddTagHelper

0.98

 themſelves

0.95

 chofe

0.92

 uſe

0.91

 becauſe

0.90

 foncé

0.90

 myſelf

0.89

 ſtand

0.86

Activations Density 0.436%

verbs indicating actions, processes, or changes

No Comments

No Known Activations