INDEX

Explanations

mentions of UCLA

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

BOOK

-0.80

odic

-0.80

DonaldTrump

-0.75

ocene

-0.71

rencies

-0.70

 wagon

-0.68

alez

-0.68

ãĥĥãĥĪ

-0.68

*/(

-0.66

Sov

-0.66

POSITIVE LOGITS

 Bruins

1.06

 Graduate

0.89

 Extension

0.86

 UCLA

0.86

 University

0.82

UC

0.81

 College

0.81

 alumni

0.79

 undergrad

0.79

 Libraries

0.79

Activations Density 0.005%

mentions of UCLA

No Comments

No Known Activations