INDEX

Explanations

expressions of negative emotions or concerns related to various issues or events

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

icle

-0.74

icles

-0.63

irrel

-0.61

ãĥ¼ãĥĨãĤ£

-0.58

Clock

-0.57

 adolesc

-0.57

 Beet

-0.56

apeake

-0.55

 Loaded

-0.55

eor

-0.54

POSITIVE LOGITS

 regarding

1.06

 towards

0.94

 toward

0.88

 sentiments

0.82

 about

0.82

 concerning

0.81

 Regarding

0.79

 expressed

0.77

 respecting

0.77

 concerns

0.76

Activations Density 0.103%

expressions of negative emotions or concerns related to various issues or events

No Comments

No Known Activations