INDEX

Explanations

mentions of monkeys

oai_token-act-pair · gpt-3.5-turbo

references to monkeys

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

 Lauder

-0.83

inen

-0.82

reek

-0.78

sburgh

-0.77

EMP

-0.75

oppable

-0.73

encer

-0.71

encers

-0.71

OHN

-0.70

aci

-0.69

POSITIVE LOGITS

pox

0.99

 wrench

0.90

sey

0.87

patch

0.82

ãĥ£

0.79

oleon

0.76

 bitten

0.73

 monkeys

0.72

zee

0.70

 monkey

0.70

Activations Density 0.021%

mentions of monkeys

references to monkeys

No Comments

No Known Activations