INDEX

Explanations

listings or items mentioned in a list

oai_token-act-pair · gpt-3.5-turbo

references to various types of listings

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

romeda

-0.76

rir

-0.76

qua

-0.72

sis

-0.71

si

-0.69

sen

-0.67

arching

-0.65

zees

-0.64

icka

-0.64

Abs

-0.64

POSITIVE LOGITS

 listings

1.37

 listing

1.31

 sellers

0.84

daq

0.80

 listed

0.76

 descriptions

0.76

 indexes

0.76

 seller

0.75

 lists

0.74

 directories

0.73

Activations Density 0.004%

listings or items mentioned in a list

references to various types of listings

No Comments

No Known Activations