INDEX

Explanations

words related to taking actions, particularly in a regulatory or enforcement context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

oran

-0.16

Å¡ov

-0.16

-0.14

vard

-0.14

orning

-0.14

 kalk

-0.14

strap

-0.14

POSITIVE LOGITS

 steps

0.33

 concrete

0.24

 Steps

0.24

 firm

0.23

 measures

0.23

steps

0.23

Steps

0.22

 appropriate

0.22

 necessary

0.21

fir

0.21

Activations Density 0.036%

words related to taking actions, particularly in a regulatory or enforcement context

No Comments

No Known Activations