INDEX

Explanations

words related to definitions or terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 photolibrary

-1.12

 pleaſure

-0.95

 myſelf

-0.92

%]

-0.90

 chofe

-0.88

elfare

-0.87

BibitemShut

-0.85

."));

-0.85

nologies

-0.85

 himſelf

-0.84

POSITIVE LOGITS

 word

2.48

Word

2.41

 Word

2.40

 WORD

2.31

word

2.20

WORD

1.93

 Words

1.57

 words

1.55

Words

1.44

 palavra

1.44

Activations Density 0.058%

words related to definitions or terminology

No Comments

No Known Activations