INDEX

Explanations

phrases related to launching campaigns or initiatives

oai_token-act-pair · gpt-3.5-turbo

phrases related to campaigns or initiatives

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

 perceived

-0.69

comes

-0.66

 outweigh

-0.64

ynes

-0.63

 fading

-0.63

nu

-0.63

 spills

-0.61

 bends

-0.60

 decay

-0.59

 sinners

-0.59

POSITIVE LOGITS

SHIP

0.73

matic

0.69

 whereby

0.67

 partnership

0.65

 titled

0.63

 crowdfunding

0.63

 Patreon

0.62

slot

0.62

study

0.62

staking

0.62

Activations Density 0.727%

phrases related to launching campaigns or initiatives

phrases related to campaigns or initiatives

No Comments

No Known Activations