INDEX

Explanations

words related to writing or marking

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to inscriptions and markings

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

kus

-0.92

EStreamFrame

-0.77

Kin

-0.76

Plug

-0.76

roxy

-0.74

rawdownloadcloneembedreportprint

-0.74

ructose

-0.71

IVERS

-0.71

bnb

-0.70

ugal

-0.67

POSITIVE LOGITS

 markings

1.18

 engraved

1.15

 inscription

1.10

 etched

1.07

prints

1.03

 inscribed

1.00

 letters

0.99

 handwritten

0.98

printed

0.97

 decipher

0.96

Activations Density 0.109%

words related to writing or marking

words and phrases related to inscriptions and markings

No Comments

No Known Activations