INDEX

Explanations

references to sharp weapons and related terms

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

ctors

-0.17

warts

-0.16

 Pill

-0.15

/boot

-0.15

osto

-0.15

 pathMatch

-0.15

-0.14

tim

-0.14

chez

-0.14

POSITIVE LOGITS

blade

0.23

 blade

0.21

-edge

0.19

(slice

0.18

 Blades

0.17

åĪĢ

0.17

 sharpen

0.17

 killings

0.16

_kill

0.16

æĿĢ

0.16

Activations Density 0.213%

references to sharp weapons and related terms

No Comments

No Known Activations