INDEX

Explanations

verbs related to actions and abilities

oai_token-act-pair · gpt-3.5-turbo

repetitive phrases related to action and intention

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĸļ

-0.66

Lenin

-0.64

Ruk

-0.63

Kub

-0.61

 Sabb

-0.54

IL

-0.54

Cly

-0.54

ruction

-0.53

 Blitz

-0.53

uggest

-0.52

POSITIVE LOGITS

 safely

0.91

 easily

0.89

ibly

0.85

ivably

0.84

 freely

0.79

 cheaply

0.77

 feas

0.75

 comfortably

0.75

nington

0.72

 anytime

0.72

Activations Density 0.337%

verbs related to actions and abilities

repetitive phrases related to action and intention

No Comments

No Known Activations