INDEX

Explanations

signs or indications of potential issues or problems

oai_token-act-pair · gpt-3.5-turbo

phrases indicating problems or concerns related to actions being taken or required

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enegger

-0.72

 Span

-0.69

iaries

-0.68

 Documents

-0.61

ilst

-0.61

 fame

-0.61

asers

-0.58

 Apostles

-0.57

aunts

-0.57

 Overs

-0.56

POSITIVE LOGITS

 wrong

1.07

 happening

0.96

wrong

0.83

 happen

0.83

 happened

0.83

 strang

0.77

 happ

0.75

 horribly

0.74

 clicked

0.71

else

0.71

Activations Density 0.134%

signs or indications of potential issues or problems

phrases indicating problems or concerns related to actions being taken or required

No Comments

No Known Activations