INDEX

Explanations

dates in the month of July

oai_token-act-pair · gpt-3.5-turbo

occurrences of the date "July 9" in the text

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

andem

-0.86

inent

-0.79

gypt

-0.79

Interstitial

-0.77

uden

-0.77

uncle

-0.76

ocrat

-0.74

¥ŀ

-0.74

 ancest

-0.72

monary

-0.71

POSITIVE LOGITS

 July

1.07

Aug

0.99

 June

0.97

Aug

0.97

June

0.95

0.94

Feb

0.94

May

0.93

July

0.91

Activations Density 0.018%

dates in the month of July

occurrences of the date "July 9" in the text

No Comments

No Known Activations