INDEX

Explanations

proper nouns or titles preceded by the word "original"

oai_token-act-pair · gpt-3.5-turbo

references to previous works or iterations

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

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

rosso

-0.83

wal

-0.80

roph

-0.76

gerald

-0.73

orst

-0.73

ega

-0.73

UTE

-0.73

robe

-0.73

BIL

-0.73

rolet

-0.72

POSITIVE LOGITS

ity

1.30

 intent

0.97

 incarnation

0.94

 intention

0.93

 trilogy

0.89

 impetus

0.89

 conception

0.88

sin

0.86

 premise

0.81

Sin

0.81

Activations Density 0.052%

proper nouns or titles preceded by the word "original"

references to previous works or iterations

No Comments

No Known Activations