INDEX

Explanations

phrases or contexts related to political discourse

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

çľł

-0.15

ATER

-0.15

rh

-0.15

ÙĨØ³

-0.14

udi

-0.14

exampleInput

-0.14

 Entire

-0.14

Ã£ng

-0.14

Ø±Ø¯Ùĩ

-0.14

amoto

-0.13

POSITIVE LOGITS

 ready

0.23

-face

0.23

IQUE

0.19

.ready

0.18

usz

0.17

issen

0.17

-NLS

0.17

ToShow

0.17

issement

0.17

 Ready

0.16

Activations Density 0.040%

phrases or contexts related to political discourse

No Comments

No Known Activations