INDEX

Explanations

verbs denoting strong emotional expressions or actions

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

 preferably

-0.68

 wisely

-0.67

 optionally

-0.67

assi

-0.66

 Advertisement

-0.66

Ki

-0.64

 equally

-0.63

 generally

-0.61

usually

-0.61

 gently

-0.60

POSITIVE LOGITS

hyde

0.67

(>

0.65

VALUE

0.64

ifiable

0.64

PsyNetMessage

0.61

anism

0.60

aeda

0.58

isable

0.58

WARD

0.58

rez

0.58

Activations Density 3.036%

verbs denoting strong emotional expressions or actions

No Comments

No Known Activations