INDEX

Explanations

numbers in the thousands

oai_token-act-pair · gpt-3.5-turbo

numerical values representing large quantities or counts in documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Cth

-0.66

sylv

-0.65

enhagen

-0.64

WAY

-0.62

yll

-0.62

ollah

-0.60

vana

-0.59

 Takeru

-0.58

uggest

-0.58

asma

-0.56

POSITIVE LOGITS

1.30

0.99

0.98

0.94

0.93

0.91

0.89

0.88

0.84

Activations Density 0.408%

numbers in the thousands

numerical values representing large quantities or counts in documents

No Comments

No Known Activations