INDEX

Explanations

words related to locations or events

oai_token-act-pair · gpt-3.5-turbo

suffixes or endings commonly associated with various nouns or adjectives

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

ãĥīãĥ©

-0.76

CLASSIFIED

-0.74

iframe

-0.73

éŃĶ

-0.72

PRES

-0.72

 pregn

-0.71

Þ

-0.69

OTOS

-0.68

TYPE

-0.68

ãĥ¤

-0.67

POSITIVE LOGITS

gow

0.97

agall

0.85

furt

0.84

opol

0.78

agles

0.77

eus

0.77

estones

0.76

amic

0.76

haven

0.74

nce

0.73

Activations Density 0.057%

words related to locations or events

suffixes or endings commonly associated with various nouns or adjectives

No Comments

No Known Activations