INDEX

Explanations

phrases that indicate methods or ways of achieving various actions or outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 Favor

-0.14

STR

-0.14

CAF

-0.13

 sacrifices

-0.13

TemplateName

-0.13

asca

-0.13

á»į

-0.13

ActionButton

-0.12

bris

-0.12

istr

-0.12

POSITIVE LOGITS

 approach

0.21

 approached

0.20

 Approach

0.20

 approaching

0.19

appro

0.18

Appro

0.18

 skin

0.17

_appro

0.17

 yaklaÅŁ

0.16

PROC

0.16

Activations Density 0.073%

phrases that indicate methods or ways of achieving various actions or outcomes

No Comments

No Known Activations