INDEX

Explanations

highlight and emphasize specific concepts or topics within longer texts or articles, often related to targets, events, or entities

oai_token-act-pair · gpt-3.5-turbo

elements related to governance and systems

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

verbs

-0.67

Parameters

-0.63

FLAG

-0.62

Lear

-0.62

à¨

-0.61

=]

-0.61

OUS

-0.61

tech

-0.59

xious

-0.59

less

-0.59

POSITIVE LOGITS

hip

1.28

cape

1.24

cale

1.22

mith

1.16

etter

1.13

heet

1.11

hips

1.10

etting

1.07

pring

1.04

pace

1.04

Activations Density 0.655%

highlight and emphasize specific concepts or topics within longer texts or articles, often related to targets, events, or entities

elements related to governance and systems

No Comments

No Known Activations