INDEX

Explanations

actions related to cutting, shaping, or modifying objects and materials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

chy

-0.17

 Tort

-0.16

 Intermediate

-0.15

ĻĤ

-0.14

asha

-0.14

 Rehab

-0.14

udes

-0.14

Uz

-0.14

scient

-0.14

okens

-0.14

POSITIVE LOGITS

 cutting

0.20

 Cutting

0.18

-cut

0.17

åĪĢ

0.17

blade

0.17

 cutter

0.16

æĪ¸

0.16

mlink

0.16

 Blasio

0.16

Sharp

0.15

Activations Density 0.145%

actions related to cutting, shaping, or modifying objects and materials

No Comments

No Known Activations