INDEX

Explanations

verbs related to decision-making and the process of making decisions

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

amon

-0.68

anto

-0.67

rongh

-0.67

acements

-0.66

icas

-0.66

oil

-0.66

eco

-0.65

attery

-0.65

udi

-0.63

atha

-0.63

POSITIVE LOGITS

 unanimously

0.94

 whether

0.84

 makers

0.80

 decision

0.79

 decisions

0.77

 calculus

0.77

 decide

0.76

 decisively

0.76

 Garc

0.76

 decided

0.73

Activations Density 1.834%

verbs related to decision-making and the process of making decisions

No Comments

No Known Activations