INDEX

Explanations

titles and references to specific media content, particularly related to "Doctor Who."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

ovice

-0.16

ÑĤÑĮ

-0.15

ibs

-0.14

ultz

-0.14

herits

-0.14

 rumored

-0.14

 stacks

-0.13

orte

-0.13

colors

-0.13

iddet

-0.13

POSITIVE LOGITS

 UNIT

0.30

 Gall

0.23

UNIT

0.21

 Torch

0.21

 Dale

0.20

 Cyber

0.20

 companion

0.19

 Companion

0.19

 companions

0.19

 sonic

0.18

Activations Density 0.006%

titles and references to specific media content, particularly related to "Doctor Who."

No Comments

No Known Activations