INDEX

Explanations

technical terms related to technology or operations

oai_token-act-pair · gpt-3.5-turbo

nouns and phrases relating to rules, events, and categorization

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

âĸ¬

-0.70

 Azerb

-0.69

 Niet

-0.66

ãĥ¼ãĥĨ

-0.60

Gaw

-0.60

Finish

-0.58

Honest

-0.57

 Palest

-0.57

athan

-0.57

Bey

-0.57

POSITIVE LOGITS

ource

1.11

peed

1.06

heet

1.04

pace

1.03

hip

1.01

paces

0.99

ilver

0.99

ystem

0.97

chool

0.96

mith

0.96

Activations Density 0.199%

technical terms related to technology or operations

nouns and phrases relating to rules, events, and categorization

No Comments

No Known Activations