INDEX

Explanations

geographical or structural references

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

inium

-0.17

 INTERRUPTION

-0.14

ITT

-0.14

 Guil

-0.14

agne

-0.14

_kw

-0.14

 forks

-0.14

esc

-0.14

 Param

-0.13

Ki

-0.13

POSITIVE LOGITS

OS

0.21

(OS

0.20

Kelly

0.20

kou

0.19

Grid

0.19

 Kelly

0.18

grid

0.18

OS

0.18

 Grid

0.18

GB

0.17

Activations Density 0.020%

geographical or structural references

No Comments

No Known Activations