INDEX

Explanations

references to the concept of duality or the presence of two elements in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

*****↵↵

-0.16

à¤¾à¤¯à¤¨

-0.16

bate

-0.16

áº¡p

-0.15

 cáº£

-0.15

iously

-0.15

ught

-0.14

 either

-0.14

 nawet

-0.14

wsz

-0.14

POSITIVE LOGITS

 sides

0.36

 sexes

0.34

 ends

0.25

/all

0.24

 genders

0.23

-sided

0.23

 halves

0.23

 sets

0.19

 parties

0.19

 Ends

0.18

Activations Density 0.067%

references to the concept of duality or the presence of two elements in various contexts

No Comments

No Known Activations