INDEX

Explanations

references to modifications or changes made to objects or systems

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

çĦ

-0.81

asar

-0.71

Ĭ

-0.69

¾

-0.69

arer

-0.68

Â¯Â¯

-0.67

ILLE

-0.67

riel

-0.66

roy

-0.66

thouse

-0.66

POSITIVE LOGITS

atile

0.86

 versions

0.86

 organisms

0.81

icum

0.79

ively

0.79

iator

0.76

ions

0.76

iations

0.75

itized

0.74

ioned

0.70

Activations Density 0.026%

references to modifications or changes made to objects or systems

No Comments

No Known Activations