INDEX

Explanations

instances of the word "seemed" followed by additional context or description

oai_token-act-pair · gpt-3.5-turbo

phrases that express perception or judgment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

por

-0.88

arta

-0.80

pora

-0.75

artisan

-0.75

opal

-0.73

cedented

-0.73

yah

-0.70

[+]

-0.68

yet

-0.67

itus

-0.67

POSITIVE LOGITS

 originally

0.93

 initially

0.92

 unsuccessfully

0.76

wolves

0.75

 formerly

0.74

 previously

0.74

 instrumental

0.71

 earlier

0.69

hes

0.68

wolf

0.65

Activations Density 0.260%

instances of the word "seemed" followed by additional context or description

phrases that express perception or judgment

No Comments

No Known Activations