INDEX

Explanations

phrases associated with time durations or deadlines

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

pour

-0.74

puff

-0.66

cart

-0.64

eri

-0.62

Desc

-0.62

ictive

-0.62

chuk

-0.60

ãĤ«

-0.60

cigarette

-0.60

 Diary

-0.60

POSITIVE LOGITS

 proven

0.85

 adulthood

0.84

 AFTER

0.77

soever

0.76

 sunrise

0.72

 midnight

0.71

 halfway

0.70

 someone

0.69

 somebody

0.68

Activations Density 0.380%

phrases associated with time durations or deadlines

No Comments

No Known Activations