INDEX

Explanations

references to highways and road-related navigation

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

ACHINE

-0.18

imore

-0.17

orthand

-0.17

dÃ¼r

-0.16

hower

-0.15

adro

-0.15

ãĥ³ãĥ

-0.15

erox

-0.15

ponible

-0.14

viron

-0.14

POSITIVE LOGITS

 dream

0.19

ot

0.17

Exc

0.15

dream

0.15

cle

0.14

arra

0.14

 Touch

0.14

 Dream

0.14

 thorough

0.14

gett

0.14

Activations Density 0.062%

references to highways and road-related navigation

No Comments

No Known Activations