INDEX

Explanations

mentions of a specific location and a governmental body

oai_token-act-pair · gpt-3.5-turbo

instances of the word "help" in contexts indicating assistance or support

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

nutrition

-0.71

 scams

-0.69

rawler

-0.69

 hatch

-0.68

avascript

-0.67

IDA

-0.66

LET

-0.65

pill

-0.65

iatus

-0.64

uits

-0.64

POSITIVE LOGITS

ħĭ

0.70

_.

0.69

 Conan

0.69

 Gret

0.66

 Quote

0.65

âĹ¼

0.64

 Analog

0.63

 Unreal

0.63

Ner

0.63

 Thousand

0.62

Activations Density 0.000%

mentions of a specific location and a governmental body

instances of the word "help" in contexts indicating assistance or support

No Comments

No Known Activations