INDEX

Explanations

terms indicating controlled substances and their characteristics

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

 Blackburn

-0.17

ahat

-0.16

ault

-0.15

oppable

-0.15

ussen

-0.15

 embryo

-0.14

umen

-0.14

aut

-0.14

ÑĩÑĸ

-0.14

fell

-0.14

POSITIVE LOGITS

 æĵ

0.14

.openConnection

0.14

 Bark

0.14

 participation

0.14

ContextHolder

0.13

rella

0.13

imer

0.13

pow

0.13

_VM

0.13

Activations Density 0.002%

terms indicating controlled substances and their characteristics

No Comments

No Known Activations