INDEX

Explanations

comparisons between the subject and others in terms of intelligence, skill, or behavior

oai_token-act-pair · gpt-3.5-turbo

expressions of comparison and self-referential remarks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Asset

-0.62

eworthy

-0.62

 moderator

-0.61

rocket

-0.60

ories

-0.59

sheet

-0.57

Dash

-0.55

LEASE

-0.55

 Shel

-0.53

Sha

-0.53

POSITIVE LOGITS

 imagined

0.87

 dreamed

0.84

 perceive

0.80

 conceive

0.78

 imagine

0.76

expected

0.75

 envisioned

0.71

 imagin

0.70

 anticipated

0.69

pect

0.69

Activations Density 0.140%

comparisons between the subject and others in terms of intelligence, skill, or behavior

expressions of comparison and self-referential remarks

No Comments

No Known Activations