INDEX

Explanations

phrases or words related to guidelines or rules

oai_token-act-pair · gpt-3.5-turbo

occurrences of the string "gu" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 swapping

-0.62

 backdrop

-0.62

 general

-0.62

rings

-0.61

 spaced

-0.61

 plank

-0.61

owl

-0.60

 Holt

-0.60

 confession

-0.59

cling

-0.59

POSITIVE LOGITS

idelines

1.67

ilty

1.54

arant

1.46

bernatorial

1.44

errilla

1.43

inea

1.38

ests

1.33

essing

1.33

inness

1.31

cci

1.30

Activations Density 0.015%

phrases or words related to guidelines or rules

occurrences of the string "gu" in various contexts

No Comments

No Known Activations