INDEX

Explanations

phrases related to something being encompassing or applicable to a wide range of situations

oai_token-act-pair · gpt-3.5-turbo

terms indicating a wide or comprehensive scope of discussion

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

CHAT

-0.64

WCS

-0.64

haven

-0.63

gers

-0.61

 Rohingya

-0.61

Absolutely

-0.61

 supreme

-0.61

aily

-0.60

Lyn

-0.60

anmar

-0.59

POSITIVE LOGITS

 suited

0.79

 developed

0.77

 behaved

0.77

 endowed

0.77

 than

0.77

 attained

0.76

 applied

0.75

utics

0.74

 formulated

0.74

 represented

0.73

Activations Density 0.115%

phrases related to something being encompassing or applicable to a wide range of situations

terms indicating a wide or comprehensive scope of discussion

No Comments

No Known Activations