INDEX

Explanations

references to locations and communities within a narrative 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

REGION

-0.18

atz

-0.16

anova

-0.15

 vign

-0.15

 putStrLn

-0.14

 pins

-0.14

oric

-0.14

reck

-0.14

 à¸¿

-0.13

ekl

-0.13

POSITIVE LOGITS

 axis

0.20

 council

0.20

 ward

0.19

 Ward

0.18

 Council

0.18

 councils

0.17

 senator

0.17

 wards

0.17

.axis

0.17

wards

0.17

Activations Density 0.023%

references to locations and communities within a narrative context

No Comments

No Known Activations