INDEX

Explanations

statements explaining reasons or causes

oai_token-act-pair · gpt-3.5-turbo

phrases that explain reasons or justifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

kees

-0.70

 torch

-0.69

odes

-0.69

apixel

-0.69

nces

-0.68

kun

-0.66

adiq

-0.66

uania

-0.65

leted

-0.63

dig

-0.63

POSITIVE LOGITS

Reason

1.13

cause

1.09

Because

1.05

 reasons

1.00

Cause

1.00

 Because

0.97

 because

0.94

ecause

0.93

because

0.87

 Reasons

0.84

Activations Density 0.216%

statements explaining reasons or causes

phrases that explain reasons or justifications

No Comments

No Known Activations