INDEX

Explanations

the name "Chip" mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Chip" in various contexts

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

 exha

-0.80

etheless

-0.74

ngth

-0.73

æĢ

-0.70

Į

-0.67

 extremes

-0.66

 ende

-0.65

 nitrogen

-0.65

 COUR

-0.64

 metab

-0.64

POSITIVE LOGITS

otle

1.67

 Chip

0.98

anut

0.96

shots

0.91

ilage

0.89

artisan

0.89

mun

0.84

unia

0.82

ambers

0.80

ards

0.78

Activations Density 0.011%

the name "Chip" mentioned in the text

mentions of the name "Chip" in various contexts

No Comments

No Known Activations