INDEX

Explanations

direct speech in quotes

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Bast

-0.62

ADS

-0.62

illon

-0.59

swick

-0.56

 hemor

-0.56

 Beacon

-0.56

 Atlantic

-0.55

 edges

-0.54

 ledger

-0.53

 latch

-0.53

POSITIVE LOGITS

'[

1.20

"[

1.16

 "âĢ¦

1.16

 "...

1.10

"@

1.09

"'

1.07

"(

1.04

""

0.98

"{

0.97

"$

0.94

Activations Density 1.756%

direct speech in quotes

No Comments

No Known Activations