INDEX

Explanations

words related to mental health conditions and experiences such as dissociation, identity confusion, and memory alteration

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

inery

-0.78

©¶æ¥µ

-0.74

abba

-0.71

 pione

-0.68

abor

-0.67

iHUD

-0.65

-0.64

alde

-0.64

ullivan

-0.63

Ãį

-0.63

POSITIVE LOGITS

1.99

?:

1.89

?'

1.86

?"

1.86

?)

1.79

?),

1.76

?",

1.75

?).

1.74

?!

1.72

?".

1.69

Activations Density 1.748%

words related to mental health conditions and experiences such as dissociation, identity confusion, and memory alteration

No Comments

No Known Activations