INDEX

Explanations

references to spiders

oai_token-act-pair · gpt-3.5-turbo

mentions of spiders

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

cussion

-0.74

brance

-0.70

ocaust

-0.69

 Lauder

-0.68

ISTER

-0.67

ufact

-0.66

endment

-0.65

ottest

-0.62

 richer

-0.61

 Commodore

-0.61

POSITIVE LOGITS

monkey

0.95

 spiders

0.95

 webs

0.95

 spider

0.93

web

0.92

lings

0.88

aceous

0.85

 silk

0.84

craft

0.81

Spider

0.81

Activations Density 0.019%

references to spiders

mentions of spiders

No Comments

No Known Activations