INDEX

Explanations

phrases indicating direction or movement

oai_token-act-pair · gpt-3.5-turbo

phrases indicating direction or movement toward a specific target or outcome

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cu

-0.80

glers

-0.72

drivers

-0.66

Ì¶

-0.66

eyes

-0.65

amp

-0.64

chin

-0.64

hai

-0.64

chu

-0.63

span

-0.63

POSITIVE LOGITS

 adulthood

0.98

 extinction

0.96

 achieving

0.96

wards

0.95

WARD

0.88

 completion

0.88

 infinity

0.87

 completing

0.86

 solving

0.83

 becoming

0.82

Activations Density 0.058%

phrases indicating direction or movement

phrases indicating direction or movement toward a specific target or outcome

No Comments

No Known Activations