INDEX

Explanations

numeric values or units mentioned in the text

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

BG

-0.72

CHR

-0.63

 Facilities

-0.60

 oils

-0.59

Cla

-0.59

IVERS

-0.59

ãĥ¼

-0.58

ende

-0.58

sbm

-0.57

Ùħ

-0.57

POSITIVE LOGITS

 apiece

1.02

undred

0.78

stroke

0.75

 accuser

0.71

secut

0.70

ousand

0.69

 lobe

0.68

fecture

0.68

icka

0.68

per

0.67

Activations Density 1.426%

numeric values or units mentioned in the text

No Comments

No Known Activations