INDEX

Explanations

statements containing the word "According."

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

IFT

-0.69

eg

-0.66

ql

-0.65

affe

-0.64

igmat

-0.63

obyl

-0.60

riage

-0.59

oppable

-0.59

estern

-0.59

apons

-0.58

POSITIVE LOGITS

 sources

0.80

SOURCE

0.73

Sources

0.70

translation

0.68

ly

0.67

edly

0.66

Ĥİ

0.66

Rank

0.65

tains

0.64

tained

0.63

Activations Density 0.355%

statements containing the word "According."

No Comments

No Known Activations