INDEX

Explanations

references to specific years or time periods

oai_token-act-pair · gpt-3.5-turbo

trends or notable changes over time

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

afety

-0.71

Fax

-0.66

 Cosponsors

-0.65

orem

-0.64

CV

-0.64

acting

-0.63

native

-0.60

Wall

-0.60

Phone

-0.60

 unconscious

-0.60

POSITIVE LOGITS

 marked

1.03

 witnessed

0.82

 tumultuous

0.80

 ausp

0.79

eco

0.75

 dominated

0.73

 characterized

0.72

 littered

0.72

 spent

0.71

ECA

0.71

Activations Density 0.251%

references to specific years or time periods

trends or notable changes over time

No Comments

No Known Activations