INDEX

Explanations

names of political figures or entities

oai_token-act-pair · gpt-3.5-turbo

prominent nouns and names related to specific people, places, or distinct entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

cially

-0.79

effective

-0.71

coded

-0.67

imately

-0.63

Effective

-0.61

orthy

-0.60

relevant

-0.60

staking

-0.60

Reloaded

-0.60

 intensive

-0.60

POSITIVE LOGITS

 citiz

0.79

 alike

0.75

 enclave

0.74

 fanbase

0.73

sylv

0.72

adan

0.70

anon

0.70

sburg

0.68

 vibe

0.67

kid

0.66

Activations Density 0.513%

names of political figures or entities

prominent nouns and names related to specific people, places, or distinct entities

No Comments

No Known Activations