INDEX

Explanations

phrases related to reasons or causes

oai_token-act-pair · gpt-3.5-turbo

phrases that denote reasons or causes behind events or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Mario

-0.71

AIN

-0.68

ena

-0.65

Italy

-0.65

forestation

-0.64

enegger

-0.64

severe

-0.64

isha

-0.64

fox

-0.63

POSITIVE LOGITS

 differences

0.89

 discrepancies

0.84

 difference

0.84

 workings

0.82

 motivations

0.79

 tendencies

0.79

 priorities

0.77

 similarities

0.77

 characteristics

0.76

 preferences

0.75

Activations Density 1.126%

phrases related to reasons or causes

phrases that denote reasons or causes behind events or actions

No Comments

No Known Activations