INDEX

Explanations

information or details following the abbreviation "i.e."

oai_token-act-pair · gpt-3.5-turbo

punctuation marks, particularly periods

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

itiz

-0.74

istar

-0.65

 polyg

-0.64

orn

-0.61

 opportunity

-0.61

 rall

-0.61

ima

-0.60

iens

-0.60

 outreach

-0.59

 burglary

-0.57

POSITIVE LOGITS

wikipedia

0.83

-)

0.76

asp

0.76

psc

0.76

sbm

0.74

+)

0.74

jpg

0.74

large

0.74

suff

0.73

Downloadha

0.72

Activations Density 0.047%

information or details following the abbreviation "i.e."

punctuation marks, particularly periods

No Comments

No Known Activations