INDEX

Explanations

references to health and safety, particularly in the context of medical care and community health issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

avia

-0.16

antages

-0.15

assembly

-0.14

igor

-0.14

ims

-0.14

 Federal

-0.14

inin

-0.14

amma

-0.14

ocus

-0.13

ump

-0.13

POSITIVE LOGITS

Outputs

0.15

Ú¯Ø§Ùĩ

0.14

reck

0.14

 lÃ¶

0.14

reed

0.14

æ¬¾

0.14

 Opposition

0.14

bbie

0.14

Mec

0.13

CI

0.13

Activations Density 0.010%

references to health and safety, particularly in the context of medical care and community health issues

No Comments

No Known Activations