INDEX

Explanations

scientific and technical terms or processes

oai_token-act-pair · gpt-3.5-turbo

phrases or terms that define or label concepts or phenomena

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

mails

-0.67

ummies

-0.66

 horrified

-0.64

udi

-0.64

 pages

-0.62

 Authorities

-0.62

 bombings

-0.61

 wiser

-0.60

SPONSORED

-0.60

 struggling

-0.60

POSITIVE LOGITS

 Passive

0.79

secondary

0.75

 epigen

0.74

phas

0.73

erred

0.72

sein

0.72

 delta

0.71

 gluc

0.69

 methyl

0.69

Activations Density 0.140%

scientific and technical terms or processes

phrases or terms that define or label concepts or phenomena

No Comments

No Known Activations