INDEX

Explanations

numerical values with decimals

oai_token-act-pair · gpt-3.5-turbo

numerical values or statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 unders

-0.71

CAM

-0.69

 outweigh

-0.65

 McGill

-0.63

FAR

-0.63

TAG

-0.63

 Opposition

-0.59

 Entered

-0.59

 Kinder

-0.59

MEM

-0.59

POSITIVE LOGITS

1.15

1.14

1.12

1.10

1.09

1.08

Activations Density 0.084%

numerical values with decimals

numerical values or statistics

No Comments

No Known Activations