INDEX

Explanations

phrases related to the "Mass Effect" video game series

oai_token-act-pair · gpt-3.5-turbo

instances of the term "Effect" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 whisk

-0.62

 legs

-0.62

 hips

-0.62

stretched

-0.58

ser

-0.58

asca

-0.57

upper

-0.57

 submitted

-0.55

 trucks

-0.55

theless

-0.54

POSITIVE LOGITS

iveness

1.25

ively

1.06

 Andromeda

0.99

ually

0.98

 Effect

0.96

enza

0.86

ual

0.84

 Awakens

0.82

ional

0.80

uality

0.80

Activations Density 0.006%

phrases related to the "Mass Effect" video game series

instances of the term "Effect" in various contexts

No Comments

No Known Activations