INDEX

Explanations

mentions of specific sections or categories within a publication

oai_token-act-pair · gpt-3.5-turbo

references to editorial selections or curated content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 nerv

-0.63

 peac

-0.58

 withdrawals

-0.58

 includ

-0.55

gal

-0.54

amen

-0.54

aten

-0.53

 foreseeable

-0.53

served

-0.52

 loudspe

-0.52

POSITIVE LOGITS

 Picks

0.94

 Latest

0.67

UGC

0.65

aukee

0.63

ļéĨĴ

0.61

heet

0.61

initions

0.60

 Getting

0.60

onga

0.59

itars

0.58

Activations Density 0.020%

mentions of specific sections or categories within a publication

references to editorial selections or curated content

No Comments

No Known Activations