INDEX

Explanations

names or terms that start with 'Ste' and 'yn'

oai_token-act-pair · gpt-3.5-turbo

names of individuals and specific entities

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

berra

-0.81

 concess

-0.79

 bluff

-0.71

ãĥīãĥ©ãĤ´ãĥ³

-0.70

ailability

-0.69

gdala

-0.67

ASP

-0.66

ILA

-0.63

 Substance

-0.63

iannopoulos

-0.62

POSITIVE LOGITS

wagen

0.83

adder

0.78

poll

0.75

furt

0.75

forth

0.72

wart

0.72

Ã«

0.71

bourg

0.71

forward

0.70

 sidx

0.70

Activations Density 0.082%

names or terms that start with 'Ste' and 'yn'

names of individuals and specific entities

No Comments

No Known Activations