INDEX

Explanations

words related to historical events or the past

oai_token-act-pair · gpt-3.5-turbo

references to the past in relation to current events or discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

starting

-0.70

 Franch

-0.70

shapeshifter

-0.63

%]

-0.61

pur

-0.60

amber

-0.59

Pick

-0.58

utm

-0.57

Hi

-0.57

alert

-0.57

POSITIVE LOGITS

ebin

1.21

ime

1.13

 tense

1.10

iche

1.05

orate

1.03

imes

1.00

decade

0.96

 millennium

0.95

 century

0.92

ures

0.86

Activations Density 0.052%

words related to historical events or the past

references to the past in relation to current events or discussions

No Comments

No Known Activations