INDEX

Explanations

statements indicating emotions or reactions in dialogue

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

ixel

-0.18

ellido

-0.16

ÈĽi

-0.16

itbart

-0.16

panion

-0.15

agr

-0.15

llib

-0.14

ëĤĺëĿ¼

-0.14

Ã³c

-0.14

opsis

-0.14

POSITIVE LOGITS

 plays

0.16

 playing

0.16

 Plays

0.15

 declar

0.15

 players

0.15

 locker

0.15

 Players

0.14

 player

0.14

 boarding

0.14

 said

0.14

Activations Density 0.100%

statements indicating emotions or reactions in dialogue

No Comments

No Known Activations