INDEX

Explanations

mentions of roles and responsibilities 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

apon

-0.18

iglia

-0.18

rael

-0.17

don

-0.16

rale

-0.16

APON

-0.16

mitt

-0.15

Mov

-0.15

rit

-0.14

fold

-0.14

POSITIVE LOGITS

-playing

0.23

playing

0.23

(Role

0.21

played

0.20

(role

0.19

.Role

0.19

 played

0.19

 ROLE

0.18

 Role

0.17

 Played

0.17

Activations Density 0.030%

mentions of roles and responsibilities in various contexts

No Comments

No Known Activations