INDEX

Explanations

terms related to technical instructions or procedures

oai_token-act-pair · gpt-3.5-turbo

phrases indicating confusion or uncertainty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

.*

-0.81

âĢ

-0.79

.—

-0.75

.–

-0.74

ÂŃ

-0.73

LGBT

-0.73

.</

-0.71

___

-0.71

âľ

-0.71

âĺ

-0.70

POSITIVE LOGITS

 realise

0.87

 learnt

0.83

 realised

0.83

 whilst

0.82

 recognise

0.81

 emphas

0.81

 organise

0.77

 Whilst

0.76

 envis

0.74

 util

0.73

Activations Density 1.383%

terms related to technical instructions or procedures

phrases indicating confusion or uncertainty

No Comments

No Known Activations