INDEX

Explanations

phrases or words related to forts

oai_token-act-pair · gpt-3.5-turbo

references to geographical locations or regions

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

natureconservancy

-0.93

interstitial

-0.85

jriwal

-0.73

adesh

-0.73

STEM

-0.69

jack

-0.68

externalActionCode

-0.66

DER

-0.65

CHAT

-0.63

Jacob

-0.63

POSITIVE LOGITS

resses

0.89

eur

0.88

 Lauderdale

0.88

fort

0.88

ress

0.88

uitous

0.84

inho

0.83

orious

0.82

oise

0.82

ieth

0.81

Activations Density 0.024%

phrases or words related to forts

references to geographical locations or regions

No Comments

No Known Activations