INDEX

Explanations

words related to judgment or evaluation

oai_token-act-pair · gpt-3.5-turbo

instances of the word "deemed" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Travels

-0.65

laughter

-0.61

 Knife

-0.60

 Opening

-0.58

 Poker

-0.58

 whirlwind

-0.58

kn

-0.56

rawdownloadcloneembedreportprint

-0.56

 Reson

-0.55

atures

-0.54

POSITIVE LOGITS

 deemed

3.39

 deem

1.91

 deems

1.90

 considered

1.69

 termed

1.53

 construed

1.50

 judged

1.48

 regarded

1.43

 perceived

1.38

 presumed

1.37

Activations Density 0.005%

words related to judgment or evaluation

instances of the word "deemed" and its variations

No Comments

No Known Activations