INDEX

Explanations

words that contain the prefix "re-"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

aa

-0.20

rey

-0.19

ae

-0.18

led

-0.17

ingly

-0.16

ÑĢÐ°Ñĩ

-0.16

dens

-0.16

ored

-0.16

ring

-0.15

POSITIVE LOGITS

du

0.21

new

0.20

ngth

0.20

trie

0.20

nts

0.20

ality

0.20

wards

0.20

ward

0.19

ly

0.19

min

0.19

Activations Density 0.012%

words that contain the prefix "re-"

No Comments

No Known Activations