INDEX

Explanations

references to structural failures or breakdowns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ropa

-0.18

lify

-0.16

UPI

-0.16

etable

-0.15

halt

-0.15

ialized

-0.15

lox

-0.15

inals

-0.15

ji

-0.15

'gc

-0.14

POSITIVE LOGITS

 into

0.23

ible

0.20

 heap

0.19

 onto

0.19

omon

0.18

 Into

0.17

 under

0.17

/exp

0.17

 INTO

0.17

under

0.17

Activations Density 0.020%

references to structural failures or breakdowns

No Comments

No Known Activations