INDEX

Explanations

phrases or sentences containing specific symbols or special characters

oai_token-act-pair · gpt-3.5-turbo

indicative phrases related to performance metrics or results

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

Vik

-0.75

uana

-0.75

 scept

-0.72

 enqu

-0.72

 organising

-0.72

ansk

-0.72

 popul

-0.72

 behaviour

-0.71

utic

-0.71

 authorised

-0.71

POSITIVE LOGITS

NBA

1.20

Players

1.13

ESPN

1.13

sports

1.12

Dallas

1.10

Game

1.05

Ble

1.04

Detroit

1.01

NFL

1.01

Football

1.01

Activations Density 0.377%

phrases or sentences containing specific symbols or special characters

indicative phrases related to performance metrics or results

No Comments

No Known Activations