INDEX

Explanations

references to the act of hearing about something

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

sis

-0.76

isoft

-0.74

ila

-0.74

igsaw

-0.73

eros

-0.71

iru

-0.69

istration

-0.69

arcity

-0.68

gradient

-0.67

redo

-0.65

POSITIVE LOGITS

 voices

1.25

 Voices

0.98

 firsthand

0.96

 footsteps

0.95

 confessions

0.94

 aloud

0.92

 whispers

0.92

 testimonies

0.90

 loud

0.89

 podcasts

0.89

Activations Density 0.485%

references to the act of hearing about something

No Comments

No Known Activations