INDEX

Explanations

photographs and related descriptions

oai_token-act-pair · gpt-3.5-turbo

references to photos and their associated details

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cffff

-0.88

conservancy

-0.79

indal

-0.76

anners

-0.75

interstitial

-0.73

paces

-0.70

transform

-0.68

adding

-0.67

artifacts

-0.66

usters

-0.63

POSITIVE LOGITS

 recipe

0.72

 source

0.70

 archived

0.66

 unknown

0.66

 john

0.64

 info

0.63

req

0.59

PM

0.59

 article

0.59

nos

0.59

Activations Density 0.151%

photographs and related descriptions

references to photos and their associated details

No Comments

No Known Activations