INDEX

Explanations

terms related to product life cycles

oai_token-act-pair · gpt-3.5-turbo

references to life or lifetimes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Monstrous

-0.73

Ô

-0.71

 Forbidden

-0.70

 Dominion

-0.67

displayText

-0.66

ItemTracker

-0.66

 Franks

-0.63

 PowerPoint

-0.63

 Awakening

-0.61

 Bernstein

-0.61

POSITIVE LOGITS

ecycle

1.39

etimes

1.35

elong

1.14

elines

1.09

estone

1.07

estones

1.07

eline

1.06

lif

1.03

lif

1.02

ework

1.00

Activations Density 0.009%

terms related to product life cycles

references to life or lifetimes

No Comments

No Known Activations