INDEX

Explanations

instances of the word "here," indicating a focus on location or presence in the text

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

GroupBox

-0.16

anco

-0.15

omid

-0.15

.paper

-0.15

ernes

-0.15

Ã©ro

-0.13

ariat

-0.13

ominated

-0.13

Ã¨me

-0.13

.proc

-0.13

POSITIVE LOGITS

ford

0.23

ina

0.20

olid

0.19

INA

0.18

lie

0.18

after

0.17

 lies

0.17

 Comes

0.17

ault

0.16

 Come

0.16

Activations Density 0.025%

instances of the word "here," indicating a focus on location or presence in the text

No Comments

No Known Activations