INDEX

Explanations

phrases related to specific locations or entities

oai_token-act-pair · gpt-3.5-turbo

specific abbreviations and acronyms

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

wel

-0.67

 "$:/

-0.66

 reckoned

-0.66

oglu

-0.65

 bounded

-0.65

 abiding

-0.64

Yel

-0.63

 VIDEOS

-0.63

bery

-0.62

 braking

-0.59

POSITIVE LOGITS

utterstock

0.89

/,

0.81

/)

0.77

RF

0.76

tenance

0.73

Lv

0.71

 alike

0.70

 Ratio

0.70

CN

0.68

avascript

0.68

Activations Density 0.148%

phrases related to specific locations or entities

specific abbreviations and acronyms

No Comments

No Known Activations