INDEX

Explanations

references to a person named "Bus" with varying levels of intensity

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Bus"

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

arily

-0.77

vironment

-0.77

vertisement

-0.74

theless

-0.70

 Pradesh

-0.67

ONY

-0.65

aneous

-0.62

 marrow

-0.61

arial

-0.61

 text

-0.60

POSITIVE LOGITS

loads

1.03

hel

0.95

Bus

0.94

bus

0.91

driver

0.91

INESS

0.84

Bus

0.84

load

0.84

boys

0.83

chu

0.81

Activations Density 0.011%

references to a person named "Bus" with varying levels of intensity

mentions of the word "Bus"

No Comments

No Known Activations