INDEX

Explanations

specifications related to colors and shapes in datasets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

pdf

-0.14

izen

-0.14

 Dude

-0.14

ForRow

-0.14

IP

-0.14

ë¨¹

-0.13

.sponge

-0.13

 Spit

-0.13

 Westbrook

-0.13

POSITIVE LOGITS

pest

0.16

 Codes

0.16

lod

0.15

Äįast

0.15

angstrom

0.15

usta

0.15

agram

0.14

_OPTS

0.14

.enum

0.14

enger

0.14

Activations Density 0.009%

specifications related to colors and shapes in datasets

No Comments

No Known Activations