INDEX

Explanations

references to snapshots or moments captured in time

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "snapshot" and its variations in different contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orney

-0.77

uct

-0.75

andi

-0.72

iant

-0.72

ricular

-0.71

ulhu

-0.70

endish

-0.67

onsense

-0.65

anguage

-0.65

ophers

-0.65

POSITIVE LOGITS

 snapshots

1.28

 snapshot

1.25

shots

1.03

snap

1.00

 Snap

0.91

Snap

0.85

shot

0.81

Shot

0.79

oleon

0.78

staking

0.73

Activations Density 0.009%

references to snapshots or moments captured in time

occurrences of the word "snapshot" and its variations in different contexts

No Comments

No Known Activations