INDEX

Explanations

place names and addresses

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

arg

-0.16

azar

-0.15

Ð°ÑĢÐ°

-0.15

au

-0.15

ÐµÐ³Ð°

-0.14

 Ø¨ÙĪØ§Ø¨Ø©

-0.14

vetica

-0.14

binary

-0.13

 anom

-0.13

beg

-0.13

POSITIVE LOGITS

ìĦł

0.18

JsonValue

0.16

_invoke

0.15

ldb

0.15

ãĢĪ

0.14

semblies

0.14

ellas

0.14

Wor

0.14

#:

0.14

_DEPRECATED

0.14

Activations Density 0.149%

place names and addresses

No Comments

No Known Activations