INDEX

Explanations

dates in the format month, day, year

oai_token-act-pair · gpt-3.5-turbo

specific years mentioned 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 @ 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

 cumbers

-0.69

avorite

-0.67

 endless

-0.64

 halves

-0.64

 heats

-0.63

 predec

-0.63

 whistle

-0.62

 numerical

-0.61

 greedy

-0.61

 intrig

-0.61

POSITIVE LOGITS

assetsadobe

0.76

 Â·

0.73

olitics

0.73

nown

0.73

)</

0.73

 âĶ

0.71

 Converted

0.71

ramid

0.69

0.68

Activations Density 0.023%

dates in the format month, day, year

specific years mentioned in the text

No Comments

No Known Activations