INDEX

Explanations

numerical patterns or sequences

oai_token-act-pair · gpt-3.5-turbo

numerical and quantifiable data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Corrections

-0.67

natureconservancy

-0.67

 FranÃ§

-0.64

 PUBLIC

-0.63

 Reborn

-0.62

 Akin

-0.62

 Elves

-0.61

accompan

-0.60

REDACTED

-0.59

 Bots

-0.58

POSITIVE LOGITS

ousand

1.06

undred

0.95

perature

0.95

velength

0.87

otted

0.85

opter

0.83

angled

0.80

secut

0.80

angle

0.80

ared

0.80

Activations Density 0.082%

numerical patterns or sequences

numerical and quantifiable data

No Comments

No Known Activations