INDEX

Explanations

time-related terms and duration 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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

untled

-0.74

illions

-0.72

alus

-0.68

urus

-0.68

ENT

-0.66

Dialogue

-0.62

BU

-0.61

UTF

-0.60

roth

-0.59

ubi

-0.59

POSITIVE LOGITS

 hiatus

1.01

 stint

0.96

 interval

0.88

 delay

0.87

 turnaround

0.86

 span

0.84

 ordeal

0.83

 period

0.81

 duration

0.81

tenance

0.81

Activations Density 0.048%

time-related terms and duration descriptions

No Comments

No Known Activations