INDEX

Explanations

sports team names, primarily focusing on football and basketball teams

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

ngth

-0.88

acles

-0.81

acle

-0.81

ocre

-0.79

egal

-0.75

alez

-0.74

yss

-0.72

ovych

-0.72

ules

-0.71

kj

-0.70

POSITIVE LOGITS

 Cubs

0.93

 Tribune

0.91

 Blackhawks

0.91

 Bears

0.83

ting

0.82

cess

0.80

 Bulls

0.79

aign

0.72

ten

0.72

boys

0.72

Activations Density 0.971%

sports team names, primarily focusing on football and basketball teams

No Comments

No Known Activations