INDEX

Explanations

technical terms related to computer programming and online services

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

olulu

-0.71

osaurs

-0.71

flies

-0.69

imore

-0.69

 ç¥ŀ

-0.67

 Tradable

-0.66

omore

-0.66

lio

-0.65

geons

-0.64

lihood

-0.63

POSITIVE LOGITS

Vs

1.18

Ds

1.14

Fs

1.10

Ms

1.10

Ns

1.10

Ps

1.09

BM

1.08

Gs

1.07

PS

1.07

TF

1.02

Activations Density 0.109%

technical terms related to computer programming and online services

No Comments

No Known Activations