INDEX

Explanations

phrases where people or groups are in agreement or have similar reactions to a situation or event

oai_token-act-pair · gpt-3.5-turbo

the word "alike" in various contexts

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

 Penal

-0.70

tein

-0.66

stra

-0.64

 Torrent

-0.64

omatic

-0.63

Ö¼

-0.62

Georg

-0.61

Planet

-0.61

=-=-

-0.59

ou

-0.59

POSITIVE LOGITS

 alike

1.35

lihood

0.93

 sexes

0.91

 minded

0.80

soever

0.79

 rejoice

0.79

 fascinated

0.73

 genders

0.72

 opposed

0.69

minded

0.69

Activations Density 0.017%

phrases where people or groups are in agreement or have similar reactions to a situation or event

the word "alike" in various contexts

No Comments

No Known Activations