INDEX

Explanations

phrases related to specific names or people with the letters "sh" in them

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

anwhile

-0.77

 numbered

-0.74

 Ajax

-0.70

etheless

-0.68

 Leth

-0.66

CAST

-0.61

 Tone

-0.60

ertodd

-0.59

OLOGY

-0.59

NEO

-0.59

POSITIVE LOGITS

apesh

1.20

ooting

1.20

apeshifter

1.19

rapnel

1.18

ackle

1.12

ifting

1.12

ivering

1.10

uffle

1.09

tml

1.07

adows

1.07

Activations Density 1.126%

phrases related to specific names or people with the letters "sh" in them

No Comments

No Known Activations