INDEX

Explanations

dates or years in historical contexts

oai_token-act-pair · gpt-3.5-turbo

references to specific years, particularly during the World War II era

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ative

-0.76

player

-0.72

atives

-0.71

 Thumbnails

-0.69

Score

-0.69

ator

-0.68

Asset

-0.67

able

-0.67

ATIVE

-0.66

itivity

-0.66

POSITIVE LOGITS

1.00

0.99

0.95

0.93

0.86

0.80

 Allied

0.80

0.77

0.76

 Reincarn

0.76

Activations Density 0.029%

dates or years in historical contexts

references to specific years, particularly during the World War II era

No Comments

No Known Activations