INDEX

Explanations

dates in December

oai_token-act-pair · gpt-3.5-turbo

instances of the word "December" and related date references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

aird

-0.69

yright

-0.66

utenberg

-0.65

htaking

-0.58

sterdam

-0.57

 Ital

-0.57

swick

-0.56

spell

-0.56

enh

-0.55

lder

-0.55

POSITIVE LOGITS

1.03

0.96

EMBER

0.96

0.95

0.94

0.92

0.90

0.87

Activations Density 0.046%

dates in December

instances of the word "December" and related date references

No Comments

No Known Activations