INDEX

Explanations

capital letter acronyms related to organizations or entities

oai_token-act-pair · gpt-3.5-turbo

references to "ML" or related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

nces

-0.80

ãĥĦ

-0.75

ricular

-0.74

forcing

-0.72

 Normandy

-0.68

lished

-0.67

ãĥĹ

-0.66

Downloadha

-0.64

liest

-0.64

 Wildcats

-0.63

POSITIVE LOGITS

anguage

0.95

TON

0.88

aughter

0.87

endon

0.87

arge

0.86

ML

0.84

yrics

0.82

iquid

0.80

DN

0.80

ibrary

0.79

Activations Density 0.020%

capital letter acronyms related to organizations or entities

references to "ML" or related terminology

No Comments

No Known Activations