INDEX

Explanations

the conjunction "and" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

acas

-0.82

osure

-0.76

ciples

-0.75

phthal

-0.72

ETS

-0.70

occup

-0.69

pec

-0.67

imilar

-0.67

osures

-0.66

iji

-0.66

POSITIVE LOGITS

 incidentally

1.14

romeda

1.10

 possibly

1.10

 optionally

1.05

 rightfully

1.04

-)

1.02

 vice

1.01

 hence

1.01

 presumably

1.00

!)

1.00

Activations Density 0.038%

the conjunction "and" in various contexts

No Comments

No Known Activations