INDEX

Explanations

references to geographical locations and port-related terminology

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

ascus

-0.16

uty

-0.15

 knot

-0.15

riter

-0.14

 Clifford

-0.14

.son

-0.13

 tarz

-0.13

wen

-0.13

inger

-0.13

POSITIVE LOGITS

 Mess

0.25

 Astros

0.24

 Drama

0.23

 Void

0.23

Void

0.23

Mess

0.23

 VOID

0.21

Arg

0.21

Lar

0.19

 Therm

0.19

Activations Density 0.014%

references to geographical locations and port-related terminology

No Comments

No Known Activations