INDEX

Explanations

time-related phrases

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

SG

-0.58

YING

-0.58

Subject

-0.58

 Digest

-0.57

Rap

-0.57

idding

-0.55

 weap

-0.55

etting

-0.54

Behind

-0.54

 Useful

-0.54

POSITIVE LOGITS

been

1.67

 been

1.54

 undergone

1.31

 gotten

1.26

 Been

1.13

 fallen

1.09

 gone

1.07

 begun

1.06

 become

1.06

 risen

1.00

Activations Density 0.673%

time-related phrases

No Comments

No Known Activations