INDEX

Explanations

words related to consequences or outcomes

oai_token-act-pair · gpt-3.5-turbo

words and phrases indicating significant consequences or implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 prefers

-0.74

 consulted

-0.73

responsive

-0.72

 motions

-0.69

 liking

-0.66

 crawling

-0.64

 likes

-0.64

 Occasionally

-0.63

tis

-0.63

accompanied

-0.63

POSITIVE LOGITS

wcs

0.78

 culmination

0.74

Merit

0.74

 vind

0.74

 legitim

0.74

 renewed

0.74

ARK

0.73

 reaff

0.73

 deepening

0.72

 acknowledgment

0.72

Activations Density 0.705%

words related to consequences or outcomes

words and phrases indicating significant consequences or implications

No Comments

No Known Activations