INDEX

Explanations

terms related to taking action or implementing methods to deal with a situation

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

eor

-0.70

BACK

-0.70

rol

-0.68

rolet

-0.65

itation

-0.64

yang

-0.63

ties

-0.61

oller

-0.60

romy

-0.60

ries

-0.60

POSITIVE LOGITS

 well

1.04

 opposed

1.00

 usual

0.98

ynchron

0.98

 soon

0.94

under

0.93

 efficiently

0.89

par

0.88

 follows

0.87

pired

0.87

Activations Density 0.125%

terms related to taking action or implementing methods to deal with a situation

No Comments

No Known Activations