INDEX

Explanations

references to current events or announcements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

 Gordon

-0.17

ingleton

-0.15

AllWindows

-0.15

ÑĢÑĮ

-0.15

elda

-0.14

 çģ«

-0.14

 away

-0.14

iedad

-0.14

Ð¼Ðµ

-0.13

enton

-0.13

POSITIVE LOGITS

exus

0.15

nee

0.15

.getExternal

0.14

 semiclass

0.14

achen

0.14

adier

0.13

peria

0.13

Ø¯ÛĮØ¯

0.13

akt

0.13

ikt

0.13

Activations Density 0.028%

references to current events or announcements

No Comments

No Known Activations