INDEX

Explanations

references to emotional or physical pain and suffering

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

edia

-0.17

'gc

-0.17

utsch

-0.16

 Sink

-0.15

-ÑĤÐ¾

-0.15

hof

-0.15

åħ°

-0.15

isation

-0.15

Å¡ka

-0.14

ization

-0.14

POSITIVE LOGITS

lessly

0.16

asher

0.16

 Pain

0.14

oven

0.14

idot

0.14

HttpException

0.14

chedulers

0.14

Ð´Ð°ÑĤ

0.13

angkan

0.13

-blood

0.13

Activations Density 0.044%

references to emotional or physical pain and suffering

No Comments

No Known Activations