INDEX

Explanations

dialogue segments and direct addresses in conversation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ãĥ¼ãĥ©

-0.17

rush

-0.16

ÃŃme

-0.16

ampion

-0.15

vers

-0.15

 McKay

-0.15

asu

-0.14

Ð°Ð½Ð¸ÑĨ

-0.14

PLEX

-0.14

POSITIVE LOGITS

eron

0.17

erin

0.15

ben

0.14

dives

0.14

gnore

0.14

emic

0.13

(OP

0.13

uster

0.13

rial

0.13

oord

0.13

Activations Density 0.002%

dialogue segments and direct addresses in conversation

No Comments

No Known Activations