INDEX

Explanations

phrases related to giving commands or making threats

oai_token-act-pair · gpt-3.5-turbo

verbs expressing intention or action in a conversation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

uably

-0.71

 strikingly

-0.70

 predominantly

-0.64

htaking

-0.63

auder

-0.62

 stereotypical

-0.62

¿½

-0.62

 Recent

-0.61

eatures

-0.60

 Harding

-0.60

POSITIVE LOGITS

 gonna

1.15

 gotta

0.97

 proceed

0.94

 obey

0.92

 alright

0.86

 wanna

0.84

 cooperate

0.82

 okay

0.82

bye

0.81

ain

0.81

Activations Density 0.525%

phrases related to giving commands or making threats

verbs expressing intention or action in a conversation

No Comments

No Known Activations