INDEX

Explanations

phrases related to past events or instances

oai_token-act-pair · gpt-3.5-turbo

references to past events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

anguage

-0.76

WAYS

-0.65

IZE

-0.65

CLUS

-0.65

izes

-0.65

shapeshifter

-0.64

UGH

-0.63

 nets

-0.63

ä¹ĭ

-0.62

 regulate

-0.62

POSITIVE LOGITS

ebin

1.16

 Past

0.99

oral

0.93

heny

0.88

inel

0.86

olini

0.85

ures

0.82

iche

0.82

elia

0.79

ure

0.77

Activations Density 0.017%

phrases related to past events or instances

references to past events

No Comments

No Known Activations