INDEX

Explanations

references to geographic locations and specific entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

aris

-0.15

Nova

-0.15

iolet

-0.14

ÑĢÐ¾ÑĪ

-0.14

orts

-0.14

 Prest

-0.14

ochen

-0.14

Ross

-0.14

gger

-0.14

åĲĪãĤıãģĽ

-0.14

POSITIVE LOGITS

 Chicago

0.21

 Illinois

0.19

Chicago

0.19

 chicago

0.18

.sz

0.17

ought

0.15

emer

0.15

ilan

0.15

ense

0.15

 kern

0.15

Activations Density 0.657%

references to geographic locations and specific entities

No Comments

No Known Activations