INDEX

Explanations

punctuation marks and their associated contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ustum

-0.14

okud

-0.13

 voks

-0.13

my

-0.13

 chances

-0.13

 blew

-0.13

Clickable

-0.13

ooke

-0.13

iore

-0.13

aphrag

-0.13

POSITIVE LOGITS

 meanwhile

0.27

 Meanwhile

0.24

 Else

0.24

Meanwhile

0.22

Else

0.20

 Later

0.20

 elsewhere

0.20

cut

0.20

 meantime

0.19

Cut

0.19

Activations Density 0.091%

punctuation marks and their associated contexts

No Comments

No Known Activations