INDEX

Explanations

phrases related to events or entities from the 1930s

oai_token-act-pair · gpt-3.5-turbo

mentions of the 1930s and related dates

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

ledged

-0.77

atives

-0.75

orney

-0.74

amin

-0.74

Kin

-0.74

hed

-0.73

ates

-0.71

iant

-0.71

lying

-0.70

atre

-0.69

POSITIVE LOGITS

0.94

0.89

é¾įå¥ĳå£«

0.87

0.85

0.83

0.82

Activations Density 0.015%

phrases related to events or entities from the 1930s

mentions of the 1930s and related dates

No Comments

No Known Activations