INDEX

Explanations

mentions of the color 'red'

oai_token-act-pair · gpt-3.5-turbo

terms related to bureaucratic or regulatory challenges

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

mercial

-0.76

BIL

-0.66

perty

-0.65

 cannabin

-0.65

torn

-0.64

 IMAGES

-0.63

atican

-0.62

alter

-0.62

internal

-0.61

################################

-0.60

POSITIVE LOGITS

otide

0.92

opsis

0.66

oxide

0.66

 flakes

0.63

 aspir

0.62

ioxide

0.62

oche

0.62

oise

0.62

 imprint

0.61

ribution

0.60

Activations Density 0.106%

mentions of the color 'red'

terms related to bureaucratic or regulatory challenges

No Comments

No Known Activations