INDEX

Explanations

quantitative information such as numbers, measurements, ages, and experiences

oai_token-act-pair · gpt-3.5-turbo

numerical representations of personal attributes or statistics

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

 inward

-0.84

vre

-0.80

 fres

-0.80

ĸļ

-0.79

uve

-0.79

ewitness

-0.78

iasm

-0.77

plin

-0.77

Reilly

-0.76

essee

-0.76

POSITIVE LOGITS

TBD

1.02

 Unknown

0.95

TBA

0.85

 Measures

0.85

???

0.82

 Detected

0.81

 CARD

0.80

 None

0.80

 Unlimited

0.78

 Approximately

0.77

Activations Density 0.084%

quantitative information such as numbers, measurements, ages, and experiences

numerical representations of personal attributes or statistics

No Comments

No Known Activations