INDEX

Explanations

actions or processes involving physical interactions or forces

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

igmatic

-0.84

redo

-0.80

ãĥĩãĤ£

-0.76

lvl

-0.76

seeing

-0.74

Ĥİ

-0.72

essim

-0.72

ortality

-0.71

OPE

-0.71

ONSORED

-0.70

POSITIVE LOGITS

 spree

1.07

redients

1.04

 technique

1.01

 noises

1.01

tons

1.01

 motions

0.99

 frenzy

0.98

 techniques

0.98

 pains

0.93

 behaviour

0.93

Activations Density 0.307%

actions or processes involving physical interactions or forces

No Comments

No Known Activations