INDEX

Explanations

activities and actions related to creating, planning, and searching

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ÂŃing

-0.19

ngen

-0.19

ingen

-0.19

sv

-0.18

ings

-0.16

ÏĥÏĦÎ±

-0.16

ableView

-0.16

ogne

-0.15

eraÃ§Ã£o

-0.15

Ã©gor

-0.15

POSITIVE LOGITS

redients

0.17

 equiv

0.16

gg

0.16

/up

0.16

REDIENT

0.16

/loading

0.15

ames

0.15

Heg

0.14

ainers

0.14

 promot

0.14

Activations Density 0.214%

activities and actions related to creating, planning, and searching

No Comments

No Known Activations