INDEX

Explanations

references to individuals named "Lewis"

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

iaries

-0.91

uses

-0.82

ĺħ

-0.79

ologies

-0.78

oat

-0.77

icago

-0.76

untu

-0.73

selves

-0.73

iary

-0.72

ortion

-0.72

POSITIVE LOGITS

 Carroll

0.98

 Hamilton

0.93

utenant

0.88

SON

0.88

hurst

0.84

cott

0.83

burg

0.81

Hamilton

0.79

ohn

0.79

 Ness

0.78

Activations Density 0.018%

references to individuals named "Lewis"

No Comments

No Known Activations