INDEX

Explanations

phrases where someone or something is being identified or identifying itself/herself/himself

oai_token-act-pair · gpt-3.5-turbo

words related to identification and self-identification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

fare

-0.85

ework

-0.77

mare

-0.69

orah

-0.67

ocracy

-0.66

hetti

-0.64

aptic

-0.63

umblr

-0.63

wake

-0.62

orld

-0.62

POSITIVE LOGITS

ively

0.78

pointers

0.76

 ourselves

0.70

 themselves

0.68

acons

0.66

ibly

0.65

 shortcomings

0.64

 weaknesses

0.64

ifiable

0.64

 identifying

0.63

Activations Density 0.043%

phrases where someone or something is being identified or identifying itself/herself/himself

words related to identification and self-identification

No Comments

No Known Activations