INDEX

Explanations

phrases related to technical specifications or product features

oai_token-act-pair · gpt-3.5-turbo

alphanumeric codes and abbreviations related to locations or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

taboola

-1.12

emort

-0.72

ioxide

-0.72

ciating

-0.70

BILITIES

-0.68

acly

-0.66

-0.65

chio

-0.65

pse

-0.62

-+-+

-0.62

POSITIVE LOGITS

inki

1.05

igans

0.72

ĸļ

0.72

lich

0.72

ournals

0.72

sap

0.71

quarters

0.71

 Abedin

0.70

nian

0.70

idan

0.70

Activations Density 0.570%

phrases related to technical specifications or product features

alphanumeric codes and abbreviations related to locations or organizations

No Comments

No Known Activations