INDEX

Explanations

words related to version control or updates

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

anova

-0.16

urum

-0.15

DMIN

-0.15

aper

-0.15

ãĥĭãĥĥãĤ¯

-0.14

DEM

-0.14

AREST

-0.14

ROM

-0.14

STM

-0.13

.progress

-0.13

POSITIVE LOGITS

Cit

0.15

ãĥ©ãĤ¯

0.15

Cit

0.15

ulings

0.15

iat

0.14

Bindable

0.14

 Sands

0.14

 Colum

0.14

EO

0.14

eview

0.14

Activations Density 0.010%

words related to version control or updates

No Comments

No Known Activations