INDEX

Explanations

mentions of the word "major"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "ior"

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

lished

-0.89

ENDED

-0.74

soDeliveryDate

-0.74

ivity

-0.71

isSpecialOrderable

-0.70

sheet

-0.69

hov

-0.66

alogue

-0.66

erness

-0.65

INESS

-0.65

POSITIVE LOGITS

gio

1.27

ior

1.14

iculture

0.91

ama

0.84

andom

0.84

ity

0.82

acy

0.79

acies

0.77

ani

0.76

ith

0.75

Activations Density 0.017%

mentions of the word "major"

occurrences of the word "ior"

No Comments

No Known Activations