INDEX

Explanations

phrases emphasizing a large quantity or a high degree

oai_token-act-pair · gpt-3.5-turbo

the word "ton."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Danger

-0.65

 Cust

-0.64

pring

-0.64

ELD

-0.63

 Forever

-0.63

 Journals

-0.63

 Psychic

-0.62

 Clinic

-0.61

 Phant

-0.61

Constructed

-0.61

POSITIVE LOGITS

neau

1.12

ne

1.02

eful

0.93

umen

0.91

Ton

0.87

ometric

0.87

 earthqu

0.87

efully

0.86

 newcom

0.85

eless

0.84

Activations Density 0.026%

phrases emphasizing a large quantity or a high degree

the word "ton."

No Comments

No Known Activations