INDEX

Explanations

the word "cut" as well as phrases related to reductions or simplifications

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Shells

-0.64

 volunte

-0.60

 Ferdinando

-0.60

 Mlle

-0.59

 Valerio

-0.59

 Când

-0.57

 Stretcher

-0.56

 Mounts

-0.56

 Illus

-0.56

 Alejandra

-0.55

POSITIVE LOGITS

cut

1.34

cut

1.30

 cuts

1.25

cutting

1.20

CUT

1.19

Cut

1.19

Cut

1.17

 cutting

1.17

cuts

1.16

CUT

1.13

Activations Density 0.103%

the word "cut" as well as phrases related to reductions or simplifications

No Comments

No Known Activations