INDEX

Explanations

information related to mental health and welfare issues, specifically focusing on search terms related to suicide, welfare eligibility, bill signing, and societal concerns

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 awoken

-0.64

.)

-0.62

exclusive

-0.60

 Canaver

-0.59

pard

-0.56

Picture

-0.54

sided

-0.54

ensional

-0.54

absolutely

-0.54

emen

-0.54

POSITIVE LOGITS

etc

1.40

!,

0.74

,...

0.71

 thereof

0.69

?,

0.68

 blah

0.66

etc

0.65

ect

0.64

Â®,

0.61

+,

0.60

Activations Density 2.803%

information related to mental health and welfare issues, specifically focusing on search terms related to suicide, welfare eligibility, bill signing, and societal concerns

No Comments

No Known Activations