INDEX

Explanations

instances of numerical references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

 Raider

-0.17

-0.15

WEST

-0.15

ock

-0.14

 Grove

-0.14

olan

-0.14

agara

-0.14

_TA

-0.14

hammad

-0.14

ala

-0.13

POSITIVE LOGITS

irit

0.16

mue

0.15

ixer

0.14

azen

0.14

ctp

0.14

ampie

0.14

plit

0.14

:Event

0.13

inspace

0.13

âĢĮØ§ÛĮ

0.13

Activations Density 0.014%

instances of numerical references

No Comments

No Known Activations