INDEX

Explanations

terms related to sensitive societal issues, particularly surrounding health and safety

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

thr

-0.17

oola

-0.15

uggy

-0.15

.Hash

-0.15

_HAL

-0.14

.ends

-0.14

éľĬ

-0.14

 discour

-0.14

èķ

-0.13

PING

-0.13

POSITIVE LOGITS

ä½ľä¸º

0.18

iver

0.16

bil

0.16

inoa

0.15

ota

0.15

strand

0.15

iew

0.14

ran

0.14

uste

0.14

ä½ľ

0.14

Activations Density 0.315%

terms related to sensitive societal issues, particularly surrounding health and safety

No Comments

No Known Activations