INDEX

Explanations

technology-related terms and specifically focuses on features or functionalities

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

sis

-0.78

oner

-0.77

alg

-0.74

 millenn

-0.72

externalActionCode

-0.72

cffffcc

-0.71

esta

-0.71

abama

-0.71

ignt

-0.69

 gloom

-0.67

POSITIVE LOGITS

ivity

0.81

hered

0.81

 features

0.81

ttes

0.78

 prominently

0.75

iveness

0.75

feature

0.75

Feature

0.75

Features

0.73

 Features

0.72

Activations Density 0.038%

technology-related terms and specifically focuses on features or functionalities

No Comments

No Known Activations