INDEX

Explanations

phrases indicating quantities

oai_token-act-pair · gpt-3.5-turbo

demonstrative pronouns and the word "these."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Held

-0.73

imaru

-0.73

-0.66

verend

-0.65

ãģª

-0.64

fred

-0.64

vey

-0.63

Sgt

-0.63

 Santana

-0.62

éļ

-0.62

POSITIVE LOGITS

outer

0.72

ierra

0.68

 succeeded

0.66

 revolves

0.64

 involved

0.63

 consisted

0.63

 succeeds

0.62

 helicop

0.62

 entails

0.62

acterial

0.62

Activations Density 0.089%

phrases indicating quantities

demonstrative pronouns and the word "these."

No Comments

No Known Activations