INDEX

Explanations

technical terms related to software coding and functionality

oai_token-act-pair · gpt-3.5-turbo

special characters and symbols used in headings or titles within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

mble

-0.79

ombo

-0.79

qus

-0.73

iot

-0.72

ilipp

-0.72

bably

-0.70

illard

-0.69

arrett

-0.66

ucket

-0.65

illin

-0.64

POSITIVE LOGITS

 Allows

0.70

 Provides

0.70

 meanwhile

0.67

 Another

0.66

 Located

0.65

":["

0.64

 Slate

0.63

 Cosponsors

0.63

 Increases

0.63

 Uses

0.62

Activations Density 0.329%

technical terms related to software coding and functionality

special characters and symbols used in headings or titles within the text

No Comments

No Known Activations