INDEX

Explanations

verbs related to making choices or 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

brance

-0.83

ijk

-0.79

TPS

-0.73

issance

-0.72

WARN

-0.71

itement

-0.70

peria

-0.70

interrupted

-0.69

warming

-0.67

naissance

-0.67

POSITIVE LOGITS

axe

0.93

 choices

0.92

 choice

0.91

 chosen

0.86

 selections

0.84

 wisely

0.84

 winners

0.83

Choice

0.80

 chooses

0.79

inations

0.79

Activations Density 1.817%

verbs related to making choices or decisions

No Comments

No Known Activations