INDEX

Explanations

mentions of the word "Gold" followed by a number

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "Gold."

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

ATIONS

-0.89

ufact

-0.70

ROR

-0.69

================================================================

-0.67

Downloadha

-0.67

uated

-0.66

========

-0.65

SPONSORED

-0.64

 srfAttach

-0.64

etheless

-0.63

POSITIVE LOGITS

smith

0.94

vertisement

0.94

finger

0.94

stone

0.93

fields

0.93

sch

0.92

mund

0.92

 medal

0.91

wyn

0.88

stein

0.87

Activations Density 0.011%

mentions of the word "Gold" followed by a number

occurrences of the term "Gold."

No Comments

No Known Activations