INDEX

Explanations

phrases or sentences comparing different entities or concepts

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing comparative structures or similarities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

umb

-0.75

 balcon

-0.68

 escal

-0.66

 unequ

-0.64

 bystand

-0.64

 flares

-0.63

 extraord

-0.62

emer

-0.62

ama

-0.61

omet

-0.60

POSITIVE LOGITS

sex

0.72

ricanes

0.70

ounter

0.67

dragon

0.67

roman

0.66

riers

0.66

aign

0.64

aneously

0.64

 Rico

0.64

same

0.64

Activations Density 0.045%

phrases or sentences comparing different entities or concepts

phrases emphasizing comparative structures or similarities

No Comments

No Known Activations