INDEX

Explanations

fragments of text related to programming or technical language

oai_token-act-pair · gpt-3.5-turbo

phrases related to structured communication or language organization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anwhile

-0.72

oppable

-0.65

romeda

-0.64

bledon

-0.63

Monitor

-0.59

 Hutch

-0.59

iership

-0.59

emis

-0.59

 funded

-0.58

 Monitoring

-0.58

POSITIVE LOGITS

 phrases

1.39

 verbs

1.28

 quotation

1.25

 phrase

1.25

 noun

1.25

 verb

1.23

phr

1.23

 ital

1.23

 adject

1.22

 words

1.21

Activations Density 0.965%

fragments of text related to programming or technical language

phrases related to structured communication or language organization

No Comments

No Known Activations