INDEX

Explanations

references to time periods, specifically focusing on spans of years

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

ople

-0.85

anooga

-0.71

eus

-0.70

ibrary

-0.69

 subp

-0.62

liction

-0.61

acly

-0.61

esc

-0.60

enburg

-0.60

abeth

-0.59

POSITIVE LOGITS

ago

1.62

Ago

1.32

 olds

1.19

old

0.96

 elapsed

0.85

 overdue

0.82

 apiece

0.80

 hindsight

0.76

ago

0.75

old

0.73

Activations Density 0.433%

references to time periods, specifically focusing on spans of years

No Comments

No Known Activations