INDEX

Explanations

terms related to American football offenses

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

bourg

-0.86

EVA

-0.74

 Clover

-0.74

 Jenner

-0.69

Leaks

-0.66

 Lama

-0.66

Prin

-0.64

çĶŁ

-0.63

 Settlement

-0.63

Mask

-0.63

POSITIVE LOGITS

 oriented

0.83

 Offense

0.78

 offense

0.78

nel

0.74

 linemen

0.73

scill

0.73

 revolving

0.73

ENSE

0.71

mble

0.71

 rotation

0.71

Activations Density 0.028%

terms related to American football offenses

No Comments

No Known Activations