INDEX

Explanations

symbols and characters used in code or data structures

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

ãĥ³ãĥĳ

-0.16

exual

-0.16

overy

-0.15

esda

-0.15

apore

-0.14

oved

-0.14

UFFIX

-0.14

----------------------------------------------------------------------

-0.14

íļį

-0.14

POSITIVE LOGITS

çħ§

0.16

ibe

0.16

abile

0.15

_INF

0.15

endi

0.15

åģ¶

0.15

_skin

0.14

ib

0.14

BC

0.14

ici

0.14

Activations Density 0.062%

symbols and characters used in code or data structures

No Comments

No Known Activations