INDEX

Explanations

references to events or actions that are pre-planned or scheduled

oai_token-act-pair · gpt-3.5-turbo

instances of the prefix "pre-."

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

loo

-0.81

Wheel

-0.76

 Slug

-0.68

Diamond

-0.68

 Heights

-0.68

Magikarp

-0.67

ãĥīãĥ©ãĤ´ãĥ³

-0.67

ibility

-0.66

mouth

-0.66

 Depth

-0.65

POSITIVE LOGITS

requisites

1.12

req

1.03

fecture

0.98

cells

0.94

hend

0.92

ppers

0.87

arr

0.87

ppy

0.84

fect

0.84

 ejac

0.84

Activations Density 0.013%

references to events or actions that are pre-planned or scheduled

instances of the prefix "pre-."

No Comments

No Known Activations