INDEX

Explanations

references to specific individuals or entities

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

Negative Logits

antan

-0.16

emit

-0.16

omu

-0.15

umo

-0.15

phan

-0.14

olest

-0.14

ved

-0.14

usan

-0.14

ume

-0.14

Nova

-0.14

POSITIVE LOGITS

 Ment

0.18

0.16

 Mans

0.16

ank

0.15

quel

0.15

GY

0.15

mer

0.14

 midd

0.14

web

0.14

 paralle

0.14

Activations Density 0.017%

references to specific individuals or entities

No Known Activations