INDEX

Explanations

medical terms related to eyesight, such as "prescribers."

oai_token-act-pair · gpt-3.5-turbo

words related to regulations or prescriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 DISTR

-0.77

ODY

-0.77

GEN

-0.76

gen

-0.73

ICA

-0.72

ANA

-0.72

ENCE

-0.71

çĲ

-0.70

éļ

-0.68

Collins

-0.68

POSITIVE LOGITS

 hours

0.91

 Hours

0.79

 ancest

0.74

 Unlock

0.72

 psychiat

0.71

 Slay

0.70

 atmosp

0.69

 Liberties

0.68

hrs

0.68

 streng

0.66

Activations Density 0.000%

medical terms related to eyesight, such as "prescribers."

words related to regulations or prescriptions

No Comments

No Known Activations