INDEX

Explanations

prepositions followed by specific nouns or entities

oai_token-act-pair · gpt-3.5-turbo

references to "the" in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

isson

-0.75

illes

-0.75

hari

-0.71

aram

-0.68

ohm

-0.67

ieve

-0.66

fman

-0.66

 Edited

-0.66

ledged

-0.64

ptions

-0.64

POSITIVE LOGITS

 sake

1.42

 foreseeable

1.05

 fledgling

0.99

 entire

0.92

 purposes

0.92

 wearer

0.90

 rest

0.87

 nation

0.85

 budding

0.85

 nascent

0.84

Activations Density 0.252%

prepositions followed by specific nouns or entities

references to "the" in various contexts

No Comments

No Known Activations