INDEX

Explanations

names of political figures

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

-0.74

or

-0.72

-0.70

but

-0.70

↵↵

-0.68

in

-0.68

to

-0.68

all

-0.67

↵

-0.66

so

-0.66

POSITIVE LOGITS

 alkoh

1.59

 kram

1.55

 Præ

1.53

 minimalis

1.52

 kosme

1.51

 karton

1.50

 kompati

1.48

 keramik

1.48

 silikon

1.47

 antik

1.46

Activations Density 0.420%

names of political figures

No Comments

No Known Activations