INDEX

Explanations

references to things being predominantly or mainly a certain way

oai_token-act-pair · gpt-3.5-turbo

instances of the word "largely" or phrases indicating extent and influence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Chaser

-0.74

 Ambassador

-0.70

anth

-0.70

yle

-0.69

atures

-0.68

yers

-0.67

 Peaks

-0.66

iry

-0.65

abad

-0.65

 Commissioners

-0.65

POSITIVE LOGITS

 reliant

0.93

 overlooked

0.91

 unchanged

0.87

 consisted

0.87

 relying

0.87

 unaffected

0.87

 overshadowed

0.87

 rely

0.86

 relied

0.86

 comprised

0.86

Activations Density 0.018%

references to things being predominantly or mainly a certain way

instances of the word "largely" or phrases indicating extent and influence

No Comments

No Known Activations