INDEX

Explanations

words related to the concept of leadership or taking charge

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

Ø§ÙĤÙĦ

-0.14

Haj

-0.13

Å¼e

-0.13

rylic

-0.13

loth

-0.13

 addCriterion

-0.13

ronym

-0.13

igan

-0.13

ucci

-0.12

rollable

-0.12

POSITIVE LOGITS

 head

1.26

 Head

1.18

head

1.15

 heads

1.13

Head

1.10

-head

1.10

 HEAD

1.04

heads

1.02

å¤´

1.00

HEAD

0.98

Activations Density 0.372%

words related to the concept of leadership or taking charge

No Comments

No Known Activations