INDEX

Explanations

phrases related to activities or events happening in the background or behind the scenes

oai_token-act-pair · gpt-3.5-turbo

references to "behind the scenes" content or discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

RIC

-0.82

fully

-0.68

nu

-0.67

tm

-0.64

ashtra

-0.61

luster

-0.60

 anew

-0.60

 msec

-0.60

 aloud

-0.59

Pwr

-0.59

POSITIVE LOGITS

 scenes

1.07

 curtain

0.93

 Scenes

0.89

 curtains

0.76

scenes

0.75

facade

0.71

 curve

0.70

scene

0.69

 wheel

0.69

 lines

0.68

Activations Density 0.093%

phrases related to activities or events happening in the background or behind the scenes

references to "behind the scenes" content or discussions

No Comments

No Known Activations