INDEX

Explanations

patterns of numbers followed by a specific word

oai_token-act-pair · gpt-3.5-turbo

phrases related to the concept of two or more items or categories

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

USA

-0.75

JM

-0.75

lov

-0.72

Inc

-0.69

manager

-0.68

Sher

-0.68

ugu

-0.67

jury

-0.67

Fed

-0.67

IAS

-0.67

POSITIVE LOGITS

 dimensional

1.17

 distinct

1.03

fold

1.00

 thirds

0.99

 pairs

0.99

 hundred

0.97

 dozen

0.97

 consecutive

0.96

 separate

0.95

 kinds

0.95

Activations Density 0.172%

patterns of numbers followed by a specific word

phrases related to the concept of two or more items or categories

No Comments

No Known Activations