INDEX

Explanations

instances where information or updates are being shared or provided

oai_token-act-pair · gpt-3.5-turbo

instances of the word "informed"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ingu

-0.77

enf

-0.66

imposed

-0.65

pmwiki

-0.63

dx

-0.61

aden

-0.61

Fan

-0.59

aho

-0.59

bers

-0.58

EG

-0.58

POSITIVE LOGITS

 informed

0.90

ingly

0.84

 assurance

0.84

 inform

0.84

soType

0.78

 antioxid

0.77

 Consent

0.77

urally

0.77

tale

0.76

ally

0.76

Activations Density 0.013%

instances where information or updates are being shared or provided

instances of the word "informed"

No Comments

No Known Activations