INDEX

Explanations

mentions of various types of inclusions or organizational entities within a broader context

oai_token-act-pair · gpt-3.5-turbo

instances of the word "including."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

iny

-0.85

bell

-0.81

iri

-0.80

ael

-0.79

ilion

-0.78

ould

-0.77

ochond

-0.76

erb

-0.75

istle

-0.74

orc

-0.73

POSITIVE LOGITS

 those

0.89

 ones

0.73

 ours

0.73

 yours

0.71

lihood

0.69

 some

0.68

 prominently

0.63

 possibly

0.63

 visits

0.63

 references

0.63

Activations Density 0.083%

mentions of various types of inclusions or organizational entities within a broader context

instances of the word "including."

No Comments

No Known Activations