INDEX

Explanations

sequential events or steps in a process

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

md

-0.63

toe

-0.62

cos

-0.61

scription

-0.60

caps

-0.60

 Presence

-0.60

ilty

-0.59

orum

-0.58

chest

-0.58

ones

-0.58

POSITIVE LOGITS

 proceeded

1.29

 proceed

1.04

 succumb

0.90

 promptly

0.89

 abruptly

0.86

 proceeds

0.85

forth

0.80

igslist

0.79

 recons

0.76

 decide

0.76

Activations Density 2.846%

sequential events or steps in a process

No Comments

No Known Activations