INDEX

Explanations

lines or phrases indicating a transition, evolution, or change in a process or situation

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

ername

-0.78

paces

-0.74

dad

-0.71

few

-0.71

chool

-0.70

bowl

-0.69

ĸļ

-0.69

rer

-0.68

terms

-0.68

gyn

-0.68

POSITIVE LOGITS

 goodies

0.82

 misinformation

0.82

 sorts

0.82

 mayhem

0.77

 doom

0.76

 activity

0.75

 sunshine

0.75

 insanity

0.74

 fame

0.73

 contr

0.73

Activations Density 3.296%

lines or phrases indicating a transition, evolution, or change in a process or situation

No Comments

No Known Activations