INDEX

Explanations

instances of punctuation or textual markers indicating quotes or citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

]]=

-0.15

ì´

-0.14

CHANT

-0.14

Ð¾Ð±Ð°

-0.14

erea

-0.14

orra

-0.13

å¤§äºº

-0.13

 interface

-0.13

.githubusercontent

-0.13

oro

-0.13

POSITIVE LOGITS

foot

0.24

 Foot

0.21

 foot

0.20

 footnote

0.19

_foot

0.18

fn

0.17

Foot

0.16

 FOOT

0.16

-foot

0.16

.ref

0.16

Activations Density 0.024%

instances of punctuation or textual markers indicating quotes or citations

No Comments

No Known Activations