INDEX

Explanations

references to the year 1990

oai_token-act-pair · gpt-3.5-turbo

references to the 1990s

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

kell

-0.75

nery

-0.69

semble

-0.67

pta

-0.67

spons

-0.65

 lightsaber

-0.65

holder

-0.64

 hypers

-0.64

hire

-0.64

cloth

-0.63

POSITIVE LOGITS

ĸļ

1.04

å¹

0.81

sburg

0.78

-'

0.78

zsche

0.74

 census

0.72

ãĤ¦ãĤ¹

0.72

cture

0.69

ãĥ³

0.67

âĸĪâĸĪ

0.66

Activations Density 0.016%

references to the year 1990

references to the 1990s

No Comments

No Known Activations

references to the year 1990

references to the 1990s

No Comments

No Known Activations