INDEX

Explanations

phrases related to technological or programming concepts

oai_token-act-pair · gpt-3.5-turbo

instances of vague or non-specific assertions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hement

-0.75

etta

-0.73

atche

-0.66

iannopoulos

-0.65

aires

-0.63

naires

-0.62

ayson

-0.61

 Viktor

-0.60

ulkan

-0.60

 estranged

-0.59

POSITIVE LOGITS

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

1.17

ÂłÂłÂłÂł

1.11

ÂłÂłÂłÂłÂłÂłÂłÂł

1.11

Posted

0.96

posted

0.94

ÂłÂł

0.91

ÂłÂłÂł

0.87

TABLE

0.81

Anyway

0.79

LAB

0.77

Activations Density 0.567%

phrases related to technological or programming concepts

instances of vague or non-specific assertions

No Comments

No Known Activations