INDEX

Explanations

terms related to heat or high intensity, especially when combined with other words or phrases

oai_token-act-pair · gpt-3.5-turbo

terms related to hot-related conditions or themes

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

yard

-0.71

 fold

-0.67

è£

-0.66

 resemb

-0.66

INAL

-0.65

iture

-0.64

axter

-0.64

emort

-0.63

 Pett

-0.62

 wiser

-0.62

POSITIVE LOGITS

oglobin

0.69

 cher

0.68

hens

0.66

 Rockets

0.66

nings

0.63

 humid

0.62

atra

0.62

terness

0.61

 lesbians

0.61

nery

0.60

Activations Density 0.258%

terms related to heat or high intensity, especially when combined with other words or phrases

terms related to hot-related conditions or themes

No Comments

No Known Activations