INDEX

Explanations

specific entities mentioned in the document, such as product names, book titles, or locations

oai_token-act-pair · gpt-3.5-turbo

elements related to specific content categories and media references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 opposite

-0.62

 mush

-0.59

``

-0.58

lit

-0.56

 comparatively

-0.54

 fertile

-0.51

 forgiving

-0.51

''

-0.51

 theirs

-0.51

 ÏĦ

-0.51

POSITIVE LOGITS

anmar

1.02

resa

0.98

odore

0.97

agascar

0.86

foundland

0.86

pherd

0.82

bidden

0.82

romeda

0.80

jamin

0.74

xiety

0.73

Activations Density 0.911%

specific entities mentioned in the document, such as product names, book titles, or locations

elements related to specific content categories and media references

No Comments

No Known Activations