INDEX

Explanations

mentions of locations or organizations

oai_token-act-pair · gpt-3.5-turbo

references to locations and associated identifiers or codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ĸļ

-0.79

©¶æ

-0.77

uyomi

-0.77

 Cosmos

-0.76

perature

-0.74

oÄŁ

-0.74

 Crunch

-0.73

raints

-0.69

rano

-0.67

ĵĺ

-0.67

POSITIVE LOGITS

UTION

1.16

IL

1.12

ibrary

1.06

DER

1.01

ADA

0.98

IK

0.97

IENT

0.96

IB

0.95

IAS

0.94

IER

0.94

Activations Density 0.004%

mentions of locations or organizations

references to locations and associated identifiers or codes

No Comments

No Known Activations