INDEX

Explanations

verbs indicating actions or processes related to creation or production

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

/remove

-0.21

heimer

-0.19

/disable

-0.18

erk

-0.18

ings

-0.17

/use

-0.16

 naken

-0.15

giy

-0.15

ignon

-0.15

POSITIVE LOGITS

ly

0.27

 thereby

0.22

 perhaps

0.20

 therein

0.19

 thus

0.17

/logging

0.17

 notamment

0.17

/loading

0.16

ss

0.16

/mod

0.15

Activations Density 0.314%

verbs indicating actions or processes related to creation or production

No Comments

No Known Activations