INDEX

Explanations

references to "front" or front-related contexts and items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

otta

-0.16

thane

-0.16

FromClass

-0.15

keit

-0.15

ephy

-0.15

ä¾¡

-0.15

Ã«n

-0.14

raison

-0.14

its

-0.14

cles

-0.14

POSITIVE LOGITS

iers

0.23

most

0.19

-facing

0.19

/back

0.18

egra

0.17

eced

0.17

sert

0.16

ech

0.16

lines

0.15

matter

0.15

Activations Density 0.041%

references to "front" or front-related contexts and items

No Comments

No Known Activations