INDEX

Explanations

phrases and variations related to repetition and repeated actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

er

-0.18

-0.17

quier

-0.15

smith

-0.15

Î½Î¿Î¼

-0.15

bell

-0.15

ugs

-0.14

erland

-0.14

rey

-0.14

 cÃ²n

-0.14

POSITIVE LOGITS

 offenders

0.20

able

0.20

 offender

0.19

ively

0.18

itious

0.18

itive

0.18

/re

0.17

/ref

0.17

/rec

0.16

 sclerosis

0.16

Activations Density 0.036%

phrases and variations related to repetition and repeated actions

No Comments

No Known Activations