INDEX

Explanations

terms related to potential possibilities or explanations

oai_token-act-pair · gpt-3.5-turbo

references to potential explanations or possibilities

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

CLASSIFIED

-0.76

neys

-0.75

bane

-0.69

ULTS

-0.69

crim

-0.69

Staff

-0.68

ceans

-0.68

cius

-0.67

dogs

-0.66

Ship

-0.66

POSITIVE LOGITS

way

1.54

 explanation

1.49

 solution

1.49

 method

1.42

 mechanism

1.31

 ways

1.31

 rationale

1.28

 answer

1.28

 scenario

1.27

 reason

1.27

Activations Density 0.309%

terms related to potential possibilities or explanations

references to potential explanations or possibilities

No Comments

No Known Activations