INDEX

Explanations

time-related terms and durations

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

acus

-0.71

orescence

-0.70

Reviewer

-0.68

ole

-0.65

ahime

-0.64

eret

-0.61

acted

-0.61

olas

-0.61

abwe

-0.61

psons

-0.59

POSITIVE LOGITS

ago

1.19

 ahead

1.05

shy

0.93

Ago

0.91

 away

0.91

 kilometers

0.90

 apart

0.90

 weeks

0.88

 kilometres

0.86

 Away

0.84

Activations Density 1.478%

time-related terms and durations

No Comments

No Known Activations