INDEX

Explanations

verbs related to actions and decision-making

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

berus

-0.76

winner

-0.74

Applic

-0.72

apo

-0.66

pletion

-0.64

fired

-0.63

éĹ

-0.61

shut

-0.60

struction

-0.59

 confir

-0.59

POSITIVE LOGITS

 themselves

1.08

 differently

0.93

 theirs

0.90

 their

0.90

 THEIR

0.80

 alike

0.74

 endlessly

0.73

ingly

0.73

 freely

0.69

 instinctively

0.68

Activations Density 0.398%

verbs related to actions and decision-making

No Comments

No Known Activations