INDEX

Explanations

references to picking or selection processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

iously

-0.19

odore

-0.16

Abs

-0.16

-fw

-0.15

é

-0.15

benh

-0.15

audi

-0.14

otion

-0.14

\Collections

-0.14

SQ

-0.14

POSITIVE LOGITS

ering

0.23

axe

0.23

les

0.23

led

0.22

lesh

0.21

ERING

0.20

ling

0.18

asso

0.18

achu

0.18

nick

0.18

Activations Density 0.010%

references to picking or selection processes

No Comments

No Known Activations