INDEX

Explanations

geographic locations and place names, particularly counties and towns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

usk

-0.14

xfff

-0.14

ogs

-0.13

Ð½ÑĸÐ²

-0.13

ONGL

-0.13

 casc

-0.13

ernet

-0.13

Mey

-0.13

(!_

-0.13

ective

-0.13

POSITIVE LOGITS

 County

0.21

 county

0.17

cent

0.17

 Ãľl

0.15

 kÃ¥

0.14

-widgets

0.14

ç¸£

0.14

County

0.14

 Circuit

0.13

Activations Density 0.052%

geographic locations and place names, particularly counties and towns

No Comments

No Known Activations