INDEX

Explanations

numeric figures or data descriptions

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

ibaba

-0.74

cyclopedia

-0.70

-0.68

artney

-0.67

izabeth

-0.66

emies

-0.65

VICE

-0.64

ription

-0.64

lopp

-0.64

 Sussex

-0.63

POSITIVE LOGITS

 skating

0.99

 prominently

0.94

heads

0.92

head

0.84

book

0.73

table

0.71

hatt

0.71

 Rasmussen

0.69

sonian

0.68

inery

0.68

Activations Density 0.726%

numeric figures or data descriptions

No Comments

No Known Activations