INDEX

Explanations

references to governmental or political entities and actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

legen

-0.16

addock

-0.14

ĥ

-0.14

ibar

-0.14

stad

-0.14

ãĥ¼ãĥĵ

-0.14

annon

-0.14

 Dragon

-0.13

otes

-0.13

readcr

-0.13

POSITIVE LOGITS

¤í

0.14

/TT

0.14

.chapter

0.14

 ControllerBase

0.14

pat

0.14

fal

0.14

UNDLE

0.14

 irre

0.14

 yard

0.14

Activations Density 0.104%

references to governmental or political entities and actions

No Comments

No Known Activations