INDEX

Explanations

noun phrases containing a specific character sequence in a nonsensical format

oai_token-act-pair · gpt-3.5-turbo

terms and references related to nouns and their grammatical usage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ACA

-0.72

JPM

-0.71

 Walmart

-0.70

FY

-0.67

 GoPro

-0.67

psey

-0.66

 Blueprint

-0.66

CPS

-0.65

 Romero

-0.65

DAQ

-0.63

POSITIVE LOGITS

 noun

1.44

 slang

1.24

 adjective

1.17

tymology

1.16

ËĪ

1.08

 verb

1.04

 verbs

1.02

 plural

0.96

adj

0.95

É

0.94

Activations Density 0.403%

noun phrases containing a specific character sequence in a nonsensical format

terms and references related to nouns and their grammatical usage

No Comments

No Known Activations