INDEX

Explanations

information related to specific terms in various languages

oai_token-act-pair · gpt-3.5-turbo

elements related to cultural or linguistic references

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

DPR

-0.70

Preview

-0.69

Chest

-0.66

Flight

-0.66

 Chemistry

-0.64

Management

-0.64

 academia

-0.64

JPM

-0.63

 Dynamics

-0.63

 Pistons

-0.63

POSITIVE LOGITS

Äĵ

1.42

Åį

1.25

É

1.19

Ã³

1.17

Ä«

1.15

Ç

1.14

Äģ

1.12

Ãł

1.10

Ä

1.08

Å«

1.07

Activations Density 0.418%

information related to specific terms in various languages

elements related to cultural or linguistic references

No Comments

No Known Activations