INDEX

Explanations

references to academic journal volumes and issue numbers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

 Fors

-0.16

roscope

-0.16

ego

-0.15

ects

-0.14

unic

-0.14

çľģ

-0.14

eci

-0.14

rox

-0.14

omic

-0.14

idata

-0.14

POSITIVE LOGITS

ë¦½

0.15

ë³Ħ

0.15

isp

0.15

/s

0.15

(issue

0.15

-addons

0.14

elman

0.14

_bindings

0.14

 Campos

0.14

Activations Density 0.015%

references to academic journal volumes and issue numbers

No Comments

No Known Activations