INDEX

Explanations

variations of the word "work" or related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

nge

-0.17

enzie

-0.16

gov

-0.16

lectic

-0.15

.pix

-0.15

ader

-0.15

 Banc

-0.14

geh

-0.14

eti

-0.14

conn

-0.14

POSITIVE LOGITS

Wor

0.23

shipping

0.21

wor

0.20

wart

0.20

kdir

0.19

SHIP

0.17

wor

0.17

.ld

0.17

ried

0.17

ktop

0.17

Activations Density 0.006%

variations of the word "work" or related terms

No Comments

No Known Activations