INDEX

Explanations

the word "Bus" followed by a single character

oai_token-act-pair · gpt-3.5-turbo

mentions of "Bus" as a key term in the text

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

arily

-0.77

vironment

-0.71

theless

-0.70

ulhu

-0.64

 Manson

-0.63

ONY

-0.63

 phosphorus

-0.62

 Pradesh

-0.61

selves

-0.60

vertisement

-0.60

POSITIVE LOGITS

INESS

1.04

hel

0.98

loads

0.93

sell

0.92

driver

0.91

ility

0.86

chu

0.84

cles

0.84

riz

0.83

ker

0.82

Activations Density 0.021%

the word "Bus" followed by a single character

mentions of "Bus" as a key term in the text

No Comments

No Known Activations