INDEX

Explanations

mentions of spa facilities

oai_token-act-pair · gpt-3.5-turbo

references to spas and related establishments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Scotia

-0.80

 Creed

-0.72

selves

-0.70

©¶æ¥µ

-0.69

è¦ļéĨĴ

-0.69

 corrid

-0.68

 Skies

-0.67

gdala

-0.67

 captcha

-0.66

otos

-0.66

POSITIVE LOGITS

enstein

0.90

eners

0.89

ENS

0.88

ener

0.86

hes

0.84

ensing

0.81

lasses

0.80

kers

0.74

spring

0.74

ulously

0.74

Activations Density 0.035%

mentions of spa facilities

references to spas and related establishments

No Comments

No Known Activations