INDEX

Explanations

proper nouns and locations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uci

-0.15

Bundle

-0.14

/Private

-0.13

Bloc

-0.13

FY

-0.13

pline

-0.13

GridColumn

-0.13

äºŃ

-0.13

&E

-0.13

etat

-0.13

POSITIVE LOGITS

-sur

0.23

ensis

0.20

 town

0.20

 city

0.20

erland

0.17

iana

0.17

 cathedral

0.17

skÃ¡

0.17

 Cathedral

0.16

 municipality

0.16

Activations Density 0.127%

proper nouns and locations

No Comments

No Known Activations