INDEX

Explanations

mentions of proper nouns related to the word 'Stone'

oai_token-act-pair · gpt-3.5-turbo

references to "Stone" and associated terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

merce

-0.82

oresc

-0.82

ership

-0.79

orescence

-0.78

olulu

-0.76

ornia

-0.72

unct

-0.70

olitan

-0.69

ores

-0.66

NCT

-0.65

POSITIVE LOGITS

hill

0.99

falls

0.96

castle

0.90

cold

0.87

works

0.85

wright

0.85

lake

0.85

fish

0.81

 Soup

0.80

 Cold

0.79

Activations Density 0.053%

mentions of proper nouns related to the word 'Stone'

references to "Stone" and associated terms

No Comments

No Known Activations