INDEX

Explanations

explanations or reasons given for events or situations

oai_token-act-pair · gpt-3.5-turbo

the term "explanation" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ymph

-0.90

illet

-0.82

pired

-0.70

inal

-0.70

oned

-0.68

mun

-0.65

ilet

-0.64

cycl

-0.64

ony

-0.64

 Gleaming

-0.63

POSITIVE LOGITS

 explanation

1.17

 explanations

1.15

WHY

1.06

why

0.98

 explan

0.96

 Explain

0.92

 explaining

0.89

why

0.88

 explain

0.88

 Explan

0.85

Activations Density 0.010%

explanations or reasons given for events or situations

the term "explanation" in various contexts

No Comments

No Known Activations