INDEX

Explanations

phrases or comments expressing positive evaluations or compliments

oai_token-act-pair · gpt-3.5-turbo

words expressing high levels of approval or positivity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

canon

-0.74

olester

-0.66

Emb

-0.63

Phase

-0.63

NetMessage

-0.61

ensor

-0.60

Downloadha

-0.60

 lest

-0.59

ilic

-0.59

ptin

-0.58

POSITIVE LOGITS

 opportunity

0.92

sword

0.87

 idea

0.84

job

0.83

 deal

0.82

 synergy

0.80

 insight

0.78

 tasting

0.78

 outdoors

0.77

 sounding

0.76

Activations Density 0.093%

phrases or comments expressing positive evaluations or compliments

words expressing high levels of approval or positivity

No Comments

No Known Activations