INDEX

Explanations

punctuation or textual markers indicating omissions or ellipses in writing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

nam

-0.16

aze

-0.15

tam

-0.15

tone

-0.14

peak

-0.14

ording

-0.14

esh

-0.14

 Wheeler

-0.14

sets

-0.14

hit

-0.14

POSITIVE LOGITS

lick

0.17

blick

0.14

zing

0.14

¹

0.14

 Slug

0.14

ging

0.14

zy

0.14

0.13

Spell

0.13

Activations Density 0.025%

punctuation or textual markers indicating omissions or ellipses in writing

No Comments

No Known Activations