INDEX

Explanations

elements related to navigation and instructions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

 Bernstein

-0.15

à¸¤

-0.14

onica

-0.14

allee

-0.14

ARRANT

-0.14

engin

-0.14

_rewrite

-0.14

ysize

-0.14

isin

-0.14

abet

-0.13

POSITIVE LOGITS

 skip

0.81

 Skip

0.74

skip

0.69

 skipping

0.68

Skip

0.68

 skipped

0.66

 bypass

0.65

 skips

0.65

 SKIP

0.60

.skip

0.60

Activations Density 0.189%

elements related to navigation and instructions

No Comments

No Known Activations