INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

text related to the term "UE" and its variations in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

imated

-0.77

rooms

-0.71

entin

-0.71

 unfocusedRange

-0.70

asses

-0.68

ãĤ¼ãĤ¦ãĤ¹

-0.65

ãĤ¢ãĥ«

-0.65

atform

-0.62

achusetts

-0.62

 approx

-0.61

POSITIVE LOGITS

UE

1.55

UES

0.96

lla

0.95

lement

0.89

ues

0.86

uve

0.85

FI

0.83

ue

0.80

BLE

0.77

nergy

0.77

Activations Density 0.006%

proper nouns

text related to the term "UE" and its variations in context

No Comments

No Known Activations