INDEX

Explanations

years in the 1920s

oai_token-act-pair · gpt-3.5-turbo

references to specific years in the 1920s and 1930s

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ledged

-0.88

yll

-0.78

atives

-0.78

osphere

-0.75

orney

-0.74

iary

-0.73

ynt

-0.72

iant

-0.72

ramer

-0.72

ops

-0.72

POSITIVE LOGITS

1.02

1.00

0.99

0.98

0.95

0.94

0.92

0.91

Activations Density 0.014%

years in the 1920s

references to specific years in the 1920s and 1930s

No Comments

No Known Activations