INDEX

Explanations

cache-related terms and concepts

oai_token-act-pair · gpt-3.5-turbo

mentions of caching concepts in a technical context

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

lihood

-0.79

borough

-0.77

hani

-0.71

 Cambod

-0.69

iates

-0.68

stad

-0.68

 Asuka

-0.68

manship

-0.67

inse

-0.67

tons

-0.67

POSITIVE LOGITS

xia

0.93

 dumps

0.86

 cache

0.85

 Cache

0.83

 Reloaded

0.77

 caches

0.75

ãĤ§

0.68

 caching

0.68

 compressor

0.67

oute

0.67

Activations Density 0.048%

cache-related terms and concepts

mentions of caching concepts in a technical context

No Comments

No Known Activations