INDEX

Explanations

assertions and conclusions supported by research findings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ARRANT

-0.16

ammen

-0.15

ester

-0.15

Ã«r

-0.14

undry

-0.14

à¹Ĥà¸¢

-0.14

Orth

-0.14

á»¯a

-0.14

gue

-0.14

_ROUND

-0.14

POSITIVE LOGITS

0.15

eon

0.15

ÙĪØ§

0.14

 sight

0.14

atom

0.14

 bench

0.14

fe

0.14

Activations Density 0.141%

assertions and conclusions supported by research findings

No Comments

No Known Activations