INDEX

Explanations

words related to post-events or conditions, such as "post-traumatic" or "post-war"

oai_token-act-pair · gpt-3.5-turbo

references to the term "post" followed by various contexts or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Qiao

-0.91

owship

-0.77

 Orchestra

-0.75

 Bett

-0.74

ï¸

-0.70

 Beir

-0.69

 Lauder

-0.68

 Faul

-0.66

 Brill

-0.66

ãĤ¤ãĥĪ

-0.66

POSITIVE LOGITS

gres

1.36

mortem

1.31

doctoral

1.28

hum

1.20

modern

1.18

natal

1.14

greSQL

1.12

graduate

1.09

cards

1.08

ulates

1.08

Activations Density 0.036%

words related to post-events or conditions, such as "post-traumatic" or "post-war"

references to the term "post" followed by various contexts or conditions

No Comments

No Known Activations