INDEX

Explanations

lists of states or countries along with their respective ratings or mentions

oai_token-act-pair · gpt-3.5-turbo

names of U.S. states and their references 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 redes

-0.85

 vibrations

-0.72

 replacements

-0.70

 meanings

-0.68

ptoms

-0.68

 reversible

-0.68

 vectors

-0.67

 disruptive

-0.67

 readings

-0.67

DragonMagazine

-0.67

POSITIVE LOGITS

 Samoa

0.98

 Guam

0.94

 Cyprus

0.93

 Costa

0.93

 Luxembourg

0.93

 Marino

0.92

 Taiwan

0.92

Tel

0.91

 Netherlands

0.91

 Philippines

0.91

Activations Density 0.164%

lists of states or countries along with their respective ratings or mentions

names of U.S. states and their references in various contexts

No Comments

No Known Activations