INDEX

Explanations

phrases related to loading and movement

oai_token-act-pair · gpt-3.5-turbo

terms related to loading and reloading processes

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

 Lauder

-0.76

von

-0.72

omen

-0.72

Ĥİ

-0.71

kos

-0.70

Ĭ±

-0.70

vette

-0.67

nov

-0.67

bold

-0.66

vous

-0.65

POSITIVE LOGITS

 loading

1.22

loading

1.07

 loader

1.02

 loads

0.96

 Load

0.96

loader

0.90

 load

0.88

 Loading

0.88

load

0.84

 docks

0.84

Activations Density 0.004%

phrases related to loading and movement

terms related to loading and reloading processes

No Comments

No Known Activations