INDEX

Explanations

names of specific individuals

oai_token-act-pair · gpt-3.5-turbo

names of prominent individuals and locations

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

door

-0.82

Down

-0.69

Charge

-0.69

ibur

-0.68

neck

-0.68

ground

-0.67

cause

-0.66

Rush

-0.66

brother

-0.66

icular

-0.64

POSITIVE LOGITS

itives

0.88

rary

0.87

opol

0.78

ilities

0.75

erald

0.74

urate

0.73

ouls

0.71

olitan

0.69

opoulos

0.69

İ

0.68

Activations Density 0.045%

names of specific individuals

names of prominent individuals and locations

No Comments

No Known Activations