INDEX

Explanations

names of specific individuals or organizations

oai_token-act-pair · gpt-3.5-turbo

names and entities related to specific organizations or individuals

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

IBLE

-0.86

xual

-0.80

entials

-0.78

ential

-0.75

encers

-0.74

ework

-0.72

Ö¼

-0.72

iments

-0.71

phrine

-0.70

drops

-0.70

POSITIVE LOGITS

kat

0.82

won

0.81

istani

0.80

lov

0.78

patrick

0.78

orea

0.77

lar

0.76

rieg

0.76

ota

0.75

pps

0.73

Activations Density 0.065%

names of specific individuals or organizations

names and entities related to specific organizations or individuals

No Comments

No Known Activations