INDEX

Explanations

words related to health, illness, sobriety, and being awake or asleep

oai_token-act-pair · gpt-3.5-turbo

terms related to health issues and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 cascade

-0.61

 Empire

-0.61

ILA

-0.60

 awards

-0.59

 unexpl

-0.57

 opacity

-0.57

 similarity

-0.56

 reconc

-0.55

 Flavoring

-0.55

 fasc

-0.55

POSITIVE LOGITS

 enough

1.08

enough

1.06

 Enough

0.85

footed

0.84

 again

0.83

abouts

0.81

 asleep

0.78

ridden

0.77

ired

0.77

wired

0.77

Activations Density 0.303%

words related to health, illness, sobriety, and being awake or asleep

terms related to health issues and conditions

No Comments

No Known Activations