INDEX

Explanations

phrases indicating familiarity or recognition

oai_token-act-pair · gpt-3.5-turbo

references to familiarity with concepts or objects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

chance

-0.79

Chance

-0.72

scoring

-0.68

hap

-0.66

mpeg

-0.65

oner

-0.64

tan

-0.64

rate

-0.64

reme

-0.63

hemy

-0.63

POSITIVE LOGITS

 familiar

1.05

iliar

0.92

isable

0.88

 faces

0.81

igan

0.81

idad

0.80

 recognizable

0.79

iable

0.77

 enough

0.77

 unfamiliar

0.76

Activations Density 0.012%

phrases indicating familiarity or recognition

references to familiarity with concepts or objects

No Comments

No Known Activations