INDEX

Explanations

phrases related to removing or eliminating something

oai_token-act-pair · gpt-3.5-turbo

phrases or expressions related to eliminating or removing unwanted elements or regulations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Correction

-0.71

format

-0.69

onse

-0.67

ebus

-0.66

Manager

-0.64

acs

-0.64

Streamer

-0.63

Forge

-0.62

facing

-0.61

POSITIVE LOGITS

 altogether

0.92

 unnecessary

0.89

 superflu

0.88

 pesky

0.80

 needless

0.78

 unwanted

0.77

 obsolete

0.75

 blight

0.74

 redundant

0.74

 incumb

0.73

Activations Density 0.149%

phrases related to removing or eliminating something

phrases or expressions related to eliminating or removing unwanted elements or regulations

No Comments

No Known Activations