INDEX

Explanations

terms related to medical procedures or conditions

oai_token-act-pair · gpt-3.5-turbo

specific endings and segments of words often related to descriptors or classifications

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

 rusty

-0.76

 glim

-0.70

 hypers

-0.63

 schooling

-0.60

 unchecked

-0.60

 Fenrir

-0.60

 spotting

-0.59

éĹ

-0.59

 downhill

-0.58

BuyableInstoreAndOnline

-0.57

POSITIVE LOGITS

dule

0.78

å£«

0.77

ctive

0.75

ruct

0.74

amation

0.74

ption

0.72

agement

0.71

mination

0.71

TOP

0.70

ember

0.70

Activations Density 0.046%

terms related to medical procedures or conditions

specific endings and segments of words often related to descriptors or classifications

No Comments

No Known Activations