INDEX

Explanations

references to official visits

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "visit"

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

corn

-0.68

Pru

-0.63

acebook

-0.61

rencies

-0.61

rpm

-0.61

Ü

-0.61

ussen

-0.60

Jou

-0.60

 pulp

-0.59

 unden

-0.59

POSITIVE LOGITS

 abroad

0.89

acion

0.86

ations

0.77

è¦ļéĨĴ

0.77

ees

0.75

PsyNetMessage

0.70

 visits

0.70

cation

0.70

 itiner

0.69

odge

0.68

Activations Density 0.066%

references to official visits

occurrences of the word "visit"

No Comments

No Known Activations