INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "gin"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Attempts

-0.68

 elapsed

-0.64

 punishment

-0.64

HHS

-0.63

 unintended

-0.62

 supervisors

-0.62

anyahu

-0.61

 plaintiff

-0.61

isites

-0.60

District

-0.59

POSITIVE LOGITS

gin

1.48

uine

1.06

forth

1.03

atin

0.94

uin

0.94

fo

0.92

Ò

0.90

etta

0.90

herer

0.87

atus

0.86

Activations Density 0.007%

proper nouns

occurrences of the substring "gin"

No Comments

No Known Activations