INDEX

Explanations

phrases or terms related to specific technical or industry-specific content, like programming languages, companies, chemicals, and sports teams

oai_token-act-pair · gpt-3.5-turbo

names of companies or organizations involved in specific industries

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

 alike

-0.60

senal

-0.58

 Rodrigo

-0.58

 Nicarag

-0.57

 Filipino

-0.57

 sequ

-0.55

 Filip

-0.54

oqu

-0.54

 primates

-0.53

Ń·

-0.53

POSITIVE LOGITS

 Pradesh

0.75

row

0.73

intel

0.73

orbit

0.72

/+

0.64

pmwiki

0.64

+(

0.64

Firstly

0.64

aults

0.63

Activations Density 0.652%

phrases or terms related to specific technical or industry-specific content, like programming languages, companies, chemicals, and sports teams

names of companies or organizations involved in specific industries

No Comments

No Known Activations